Techniques for reporting relevance metrics for machine learning reproducibility

Abstract

Methods, systems, and devices for wireless communication are described. A machine learning server may generate a set of low-dimensional parameters representing training data for the machine learning server, the training data associated with one or more communication environments or one or more channel environments, or a combination thereof. The machine learning server may receive, from one or more devices within a communication environment or a channel environment or both, a set of low-dimensional parameters representing test data associated with the communication environment or the channel environment or both. The machine learning server may generate a reproducibility metric according to a correlation between the set of parameters representing the training data and the set of parameters representing the test data. The machine learning server may send a message indicating the reproducibility metric to the one or more devices, and the one or more devices may perform a communication procedure based on the reproducibility metric.

Description

[0001] Cross-references

[0002] This patent application claims priority to U.S. patent application No. 18 / 045,614, entitled “TECHNIQUESFOR REPORTING CORRELATION METRICS FOR MACHINE LEARNING REPRODUCIBILITY,” filed by RAGHAVAN et al. on October 11, 2022, which is assigned to the assignee of this application and is expressly incorporated herein by reference in its entirety. Technical Field

[0003] The following relates to wireless communications, including techniques for reporting correlation metrics for machine learning reproducibility. Background Art

[0004] Wireless communication systems are widely deployed to provide various types of communication content, such as voice, video, packet data, messaging, broadcast, etc. These systems may be able to support communication with multiple users by sharing available system resources (e.g., time, frequency, and power). Examples of such multiple access systems include fourth generation (4G) systems (such as long term evolution (LTE) systems, advanced LTE (LTE-A) systems, or LTE-A Pro systems) and fifth generation (5G) systems (which may be referred to as new radio (NR) systems). These systems may employ techniques such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal FDMA (OFDMA), or discrete Fourier transform spread orthogonal frequency division multiplexing (DFT-S-OFDM). A wireless multiple access communication system may include one or more base stations, each of which supports wireless communications for communication devices, which may be referred to as user equipment (UE). Summary of the invention

[0005] The described technology relates to improved methods, systems, devices, and apparatuses that support technology for reporting correlation metrics for machine learning reproducibility. For example, the described technology provides a machine learning server to generate a parameter set (e.g., a low-dimensional parameter set) representing training data used by the machine learning server, the training data being associated with one or more communication environments or one or more channel environments or a combination thereof. The machine learning server may receive a parameter set (e.g., another low-dimensional parameter set) representing test data associated with the communication environment or the channel environment or both from one or more devices within an environment (e.g., a communication environment or a channel environment or both). The machine learning server may generate a reproducibility metric based on the correlation between the parameter set representing the training data and the received parameter set representing the test data. The reproducibility metric may be associated with a confidence value of a prediction generated by the machine learning server. The machine learning server may send a message indicating the reproducibility metric to the one or more devices.

[0006] Upon receiving a message indicating a reproducibility metric, a device (e.g., a user equipment (UE), a network node) may determine whether to utilize information and predictions provided by a machine learning server to communicate in a communication environment or a channel environment. For example, where the reproducibility metric satisfies a threshold, there may be some correlation between a set of parameters representing training data and a set of parameters representing test data, which may enable relatively accurate machine learning-based predictions for determining communication parameters in a communication or channel environment. In other examples, if the reproducibility metric does not satisfy the threshold, one or more devices located in the communication or channel environment may instead rely on their own measurements and / or estimates to determine the communication parameters (e.g., until retraining is performed to update the training data).

[0007] An apparatus for wireless communication at a machine learning server is described. The apparatus may include a processor, a memory coupled to the processor, and instructions stored in the memory. The instructions may be executable by the processor to cause the apparatus to: generate a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data; receive one or more messages from one or more devices indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; and send a message indicating a reproducibility metric of the communication environment or the channel environment or both to the one or more devices, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0008] A method for wireless communication at a machine learning server is described. The method may include: generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data; receiving one or more messages from one or more devices indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; and sending a message to the one or more devices indicating a reproducibility metric for the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0009] Another apparatus for wireless communication at a machine learning server is described. The apparatus may include: a component for generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data; a component for receiving one or more messages from one or more devices, the one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; and a component for sending a message to the one or more devices, the message indicating a reproducibility metric for the communication environment or the channel environment or both, the reproducibility metric based on a correlation between the first set of parameters and the second set of parameters.

[0010] A non-transitory computer-readable medium storing code for wireless communication at a machine learning server is described. The code may include instructions that can be executed by a processor to: generate a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data; receive one or more messages from one or more devices indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; and send a message to the one or more devices indicating a reproducibility metric for the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0011] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, a reproducibility metric instructs the one or more devices to perform one or more communication procedures based on one or more predictions from a machine learning model generated by a machine learning server using training data, the reproducibility metric instructing to perform the one or more communication procedures based on the reproducibility metric satisfying a threshold.

[0012] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, a reproducibility metric indicates to the one or more devices to perform one or more communication procedures based on one or more measurements performed by the one or more devices, the reproducibility metric indicating to perform the one or more communication procedures based on the reproducibility metric failing to meet a threshold.

[0013] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, a reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by a machine learning server using training data applied to a communication environment.

[0014] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, predictions from the machine learning model generated by the machine learning server include beam weight predictions for a non-blocked scenario or a blocked scenario.

[0015] Some examples of the methods, apparatuses, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for generating a reproducibility metric based on a correlation between the first set of parameters and the second set of parameters.

[0016] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, generating a reproducibility metric may include operations, features, components, or instructions for generating a reproducibility metric based on a correlation between one or more principal components associated with training data and one or more principal components associated with test data, wherein a first set of parameters includes the one or more principal components associated with the training data, and wherein a second set of parameters includes the one or more principal components associated with the test data.

[0017] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for generating additional training data for the one or more communication environments or the one or more channel environments, or a combination thereof based on the reproducibility metric failing to meet the threshold.

[0018] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the training data includes an indication of at least one of: the number of clusters in the one or more communication environments or the one or more channel environments or a combination thereof, the number of reflectors in the one or more communication environments or the one or more channel environments or a combination thereof, the number of diffraction points in the one or more communication environments or the one or more channel environments or a combination thereof, positioning information associated with one or more devices in the one or more communication environments or the one or more channel environments or a combination thereof, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capabilities of the one or more devices, or the beam measurement capabilities of the one or more devices.

[0019] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the test data includes an indication of at least one of: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the one or more devices, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capabilities of the one or more devices, or the beam measurement capabilities of the one or more devices.

[0020] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the machine learning server can be part of a UE or part of a network entity, or include an independent node associated with a collection of multiple devices in a network.

[0021] An apparatus for wireless communication at a first device is described. The apparatus may include a processor, a memory coupled to the processor, and instructions stored in the memory. The instructions may be executable by the processor to cause the apparatus to: generate a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; send a first message indicating the second set of parameters representing the test data to a machine learning server; and receive a second message indicating a reproducibility measure based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0022] A method for wireless communication at a first device is described. The method may include: generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; sending a first message indicating the second set of parameters representing the test data to a machine learning server; and receiving a second message indicating a reproducibility measure based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0023] Another apparatus for wireless communication at a first device is described. The apparatus may include: a component for generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; a component for sending a first message to a machine learning server, the first message indicating the second set of parameters representing the test data; and a component for receiving a second message indicating a reproducibility measure based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0024] A non-transitory computer-readable medium storing code for wireless communication at a first device is described. The code may include instructions that can be executed by a processor to: generate a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; send a first message indicating the second set of parameters representing the test data to a machine learning server; and receive a second message indicating a reproducibility measure based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0025] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, a reproducibility metric indicates that one or more communication procedures are to be performed based on one or more predictions from a machine learning model generated by a machine learning server using training data, and the reproducibility metric indicates that a first device performs the one or more communication procedures based on the reproducibility metric satisfying a threshold.

[0026] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, a reproducibility metric indicates that one or more communication procedures are performed based on one or more measurements performed by a first device, the reproducibility metric indicating that the first device performs the one or more communication procedures based on the reproducibility metric failing to meet a threshold.

[0027] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, a reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by a machine learning server using training data applied to a communication environment or a channel environment, or both.

[0028] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, predictions from a machine learning model associated with training data generated by a machine learning server include beam weight predictions for a non-blocked scenario or a blocked scenario.

[0029] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for the following actions: receiving a message indicating one or more parameters from a second device within the communication environment or the channel environment or both, wherein generating the second set of parameters may be based on receiving the message indicating the one or more parameters.

[0030] Some examples of the methods, apparatuses, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for sending a message indicating the reproducibility metric to a second device within the communication environment or the channel environment or both based on receiving the reproducibility metric.

[0031] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the test data includes an indication of at least one of: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the first device, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the first device, or the beam measurement capability of the first device.

[0032] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the training data includes an indication of at least one of: the number of clusters in the one or more communication environments or the one or more channel environments or a combination thereof, the number of reflectors in the one or more communication environments or the one or more channel environments or a combination thereof, the number of diffraction points in the one or more communication environments or the one or more channel environments or a combination thereof, positioning information associated with one or more devices in the one or more communication environments or the one or more channel environments or a combination thereof, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capabilities of the one or more devices, or the beam measurement capabilities of the one or more devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 An example of a wireless communication system supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0034] Figure 2 An example of a wireless communication system supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0035] Figure 3 An example of a machine learning process supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0036] Figure 4 An example of a process flow supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0037] Figure 5 and Figure 6 A block diagram illustrating a device supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0038] Figure 7 A block diagram of a communications manager supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0039] Figure 8 A diagram illustrating a system including a UE supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0040] Fig. 9 A diagram illustrating a system including network entities supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated.

[0041] Figures 10 to 13 A flowchart illustrating a method of supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated. DETAILED DESCRIPTION

[0042] In some specific implementations of wireless communication systems that can support machine learning techniques, the output of the machine learning process (e.g., output function, prediction) can be generated based on training data. The training data can be associated with one or more scenarios for predicting beams or other communication parameters, including power saving, blocking mitigation, etc. The training data can accordingly include one or more parameter sets associated with multiple communication environments or channel environments or a combination thereof. After establishing the training data, one or more devices can collect and report test data associated with a specific environment (e.g., a test environment), which can have a large number of characteristics that affect wireless communication within the environment. For example, different test environments may include objects of different numbers and sizes that reflect, block, or otherwise affect wireless signals. For example, a crowded urban area may be associated with communication and / or channel conditions that are different from those in rural areas. In any case, the device can report test data corresponding to various environments and / or channels, which can then be used as input to the machine learning process to predict various aspects of communication for various environments and channels.

[0043] However, in some cases, the training data may not be a good representation of the specific environment in which the machine learning technique is implemented. That is, although the training data may be based on one or more communication environments or one or more channel environments or a combination thereof, there may be communication environments or channel environments that are different (e.g., significantly different) from the multiple communication environments associated with the training data. For example, the communication environment may have different numbers and positioning of reflectors, clusters, diffraction points, UEs, network entities, or other objects within the communication environment. Additionally, the antenna configuration, antenna arrangement, number of antennas, beamforming capabilities, and beam measurement capabilities may be different for one or more devices within the communication environment. When the test data is different from the training data, the machine learning technique may be abandoned entirely, possibly requiring the device to fall back to other techniques until, for example, the machine learning server can be retrained to obtain additional training data, which may be inefficient and fail to achieve the advantages of machine learning-based predictions.

[0044] According to the example as described herein, the machine learning server may generate a low-dimensional parameter set representing the training data. Similarly, one or more devices of the communication environment may generate a low-dimensional parameter set associated with the test data (e.g., data associated with the communication environment or channel environment). A low-dimensional parameter set (e.g., a low-dimensional data set) may refer to a set of parameters or data with a relatively reduced dimension relative to the original corresponding data set, which may include a relatively large amount of data, features, and / or parameters. For example, a low-dimensional parameter set representing the training data may include data with a reduced dimension relative to the dimension of the training data (e.g., data in a low-dimensional space), so that the parameter set retains some meaningful attributes of the training data. Similarly, a low-dimensional parameter set representing the test data may include data with a reduced dimension relative to the dimension of the test data. Various techniques including, for example, linear dimensionality reduction techniques, nonlinear dimensionality reduction techniques, feature extraction techniques, feature selection techniques, etc. may be used to generate a low-dimensional parameter set (e.g., data in a low-dimensional space reflecting the characteristics of data in a high-dimensional space) from the original data.

[0045] The one or more devices may send a message indicating a low-dimensional parameter set representing the test data to the machine learning server. The machine learning server may generate a reproducibility metric based on a correlation between the low-dimensional parameter set representing the training data and the received low-dimensional parameter set representing the test data. The reproducibility metric may be associated with a confidence value for the prediction generated by the machine learning server. The machine learning server may then send a message indicating the reproducibility metric to the one or more devices.

[0046] Thus, the one or more devices may perform communication procedures based on the received reproducibility metric. For example, if the reproducibility metric meets the threshold, the one or more devices may continue to rely on machine learning techniques, which may enable the device to benefit from machine learning predictions and avoid discarding such techniques entirely. That is, the reproducibility metric may indicate that there is a "good enough" correlation between the test data and the training data to utilize the machine learning technique, which may avoid having to wait for retraining before using such technique in a particular environment. In other examples, if the reproducibility metric fails to meet the threshold, the one or more devices may perform communication procedures based on measurements performed by the one or more devices (e.g., the one or more devices may fall back to traditional techniques for determining communication parameters).

[0047] Various aspects of the disclosure are first described in the context of wireless communication systems. Additionally, various aspects of the disclosure are described with reference to machine learning diagrams. Various aspects of the disclosure are further illustrated by apparatus diagrams, system diagrams, and flow diagrams relating to techniques for reporting correlation metrics for machine learning reproducibility, and various aspects of the disclosure are described with reference to these diagrams.

[0048] Figure 1 An example of a wireless communication system 100 that supports techniques for reporting correlation metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated. The wireless communication system 100 may include one or more network entities 105, one or more UEs 115, and a core network 130. In some examples, the wireless communication system 100 may be a Long Term Evolution (LTE) network, an Advanced LTE (LTE-A) network, an LTE-A Pro network, a New Radio (NR) network, or a network operating according to other systems and radio technologies, including future systems and radio technologies not explicitly mentioned herein.

[0049] The network entities 105 may be dispersed throughout a geographic area to form the wireless communication system 100, and may include devices in different forms or with different capabilities. In various examples, the network entities 105 may be referred to as network elements, mobility elements, radio access network (RAN) nodes, or network equipment, among other nomenclature. In some examples, the network entities 105 and the UE 115 may communicate wirelessly via one or more communication links 125 (e.g., radio frequency (RF) access links). For example, the network entity 105 may support a coverage area 110 (e.g., a geographic coverage area) within which the UE 115 and the network entity 105 may establish one or more communication links 125. The coverage area 110 may be an example of a geographic area within which the network entity 105 and the UE 115 may support signal communications according to one or more radio access technologies (RATs).

[0050] The UEs 115 may be dispersed throughout the coverage area 110 of the wireless communication system 100, and each UE 115 may be stationary or mobile or both at different times. The UEs 115 may be devices in different forms or with different capabilities. Figure 1 Some example UEs 115 are illustrated in FIG. 1. The UEs 115 described herein may be capable of supporting communication with various types of devices, such as Figure 1 Communicate with other UEs 115 or network entities 105) as shown.

[0051] As described herein, a node of the wireless communication system 100 (which may be referred to as a network node or a wireless node) may be a network entity 105 (e.g., any network entity described herein), a UE 115 (e.g., any UE described herein), a network controller, an apparatus, a device, a computing system, one or more components, or another suitable processing entity configured to perform any of the techniques described herein. For example, the node may be a UE 115. As another example, the node may be a network entity 105. As another example, a first node may be configured to communicate with a second node or a third node. In one aspect of this example, the first node may be a UE 115, the second node may be a network entity 105, and the third node may be a UE 115. In another aspect of this example, the first node may be a UE 115, the second node may be a network entity 105, and the third node may be a network entity 105. In other aspects of this example, the first node, the second node, and the third node may be different relative to these examples. Similarly, references to UE 115, network entity 105, apparatus, device, computing system, etc. may include disclosure of UE 115, network entity 105, apparatus, device, computing system, etc. as nodes. For example, a disclosure that UE 115 is configured to receive information from network entity 105 also discloses that the first node is configured to receive information from a second node.

[0052] In some examples, the network entities 105 may communicate with the core network 130, or with each other, or both. For example, the network entities 105 may communicate with the core network 130 via one or more backhaul communication links 120 (e.g., according to S1, N2, N3, or other interface protocols). In some examples, the network entities 105 may communicate with each other directly (e.g., directly between the network entities 105) or indirectly (e.g., via the core network 130) via the backhaul communication links 120 (e.g., according to X2, Xn, or other interface protocols). In some examples, the network entities 105 may communicate with each other via midhaul communication links 162 (e.g., according to a midhaul interface protocol) or fronthaul communication links 168 (e.g., according to a fronthaul interface protocol) or any combination thereof. The backhaul communication links 120, midhaul communication links 162, or fronthaul communication links 168 may be or include one or more wired links (e.g., electrical links, optical fiber links), one or more wireless links (e.g., radio links, wireless optical links), etc. or various combinations thereof. UE 115 may communicate with core network 130 via communication link 155 .

[0053] One or more of the network entities 105 described herein may include or may be referred to as a base station 140 (e.g., a transceiver base station, a radio base station, an NR base station, an access point, a radio transceiver, a Node B, an evolved Node B (eNB), a next generation Node B, or a Gigabit Node B (any of which may be referred to as a gNB), a 5G NB, a next generation eNB (ng-eNB), a Home Node B, a Home Evolved Node B, or other suitable terms). In some examples, the network entity 105 (e.g., a base station 140) may be implemented in an aggregated (e.g., monolithic, stand-alone) base station architecture, which may be configured to utilize a protocol stack physically or logically integrated within a single network entity 105 (e.g., a single RAN node, such as a base station 140).

[0054] In some examples, the network entity 105 may be implemented in a disaggregated architecture (e.g., a disaggregated base station architecture, a disaggregated RAN architecture) that may be configured to utilize a protocol stack that is physically or logically distributed between two or more network entities 105, such as an integrated access backhaul (IAB) network, an open RAN (O-RAN) (e.g., a network configuration sponsored by the O-RAN Alliance), or a virtualized RAN (vRAN) (e.g., a cloud RAN (C-RAN)). For example, the network entity 105 may include one or more of the following: a central unit (CU) 160, a distributed unit (DU) 165, a radio unit (RU) 170, a RAN intelligent controller (RIC) 175 (e.g., a near real-time RIC (near RT RIC), a non-real-time RIC (non-RT RIC)), a service management and orchestration (SMO) 180 system, or any combination thereof. The RU 170 may also be referred to as a radio head, a smart radio head, a remote radio head (RRH), a remote radio unit (RRU), or a transmit receive point (TRP). One or more components of the network entity 105 in the decomposed RAN architecture may be co-located, or one or more components of the network entity 105 may be located in distributed locations (e.g., separate physical locations). In some examples, one or more network entities 105 of the decomposed RAN architecture may be implemented as virtual units (e.g., virtual CU (VCU), virtual DU (VDU), virtual RU (VRU)).

[0055] The functional split between CU 160, DU 165, and RU 170 is flexible and may support different functionalities, depending on which functions (e.g., network layer functions, protocol layer functions, baseband functions, RF functions, and any combination thereof) are performed at CU 160, DU 165, or RU 170. For example, a functional split of a protocol stack may be employed between CU 160 and DU 165 such that CU 160 may support one or more layers of a protocol stack and DU 165 may support one or more different layers of a protocol stack. In some examples, CU 160 may host higher protocol layer (e.g., Layer 3 (L3), Layer 2 (L2)) functionality and signaling (e.g., Radio Resource Control (RRC), Service Data Adaptation Protocol (SDAP), Packet Data Convergence Protocol (PDCP)). The CU 160 may be connected to one or more DUs 165 or RUs 170, and the one or more DUs 165 or RUs 170 may host lower protocol layers, such as layer 1 (L1) (e.g., physical (PHY) layer) or L2 (e.g., radio link control (RLC) layer, medium access control (MAC) layer) functionality and signaling, and may each be at least partially controlled by the CU 160. Additionally or alternatively, a functional split of the protocol stack may be employed between the DU 165 and the RU 170, such that the DU 165 may support one or more layers of the protocol stack, and the RU 170 may support one or more different layers of the protocol stack. The DU 165 may support one or more different cells (e.g., via one or more RUs 170). In some cases, the functional split between CU 160 and DU 165 or between DU 165 and RU 170 may be within a protocol layer (e.g., some functions of a protocol layer may be performed by one of CU 160, DU 165, or RU 170, while other functions of the protocol layer are performed by a different one of CU 160, DU 165, or RU 170). CU 160 may be further functionally split into CU control plane (CU-CP) and CU user plane (CU-UP) functions. CU 160 may be connected to one or more DUs 165 via midhaul communication links 162 (e.g., F1, F1-c, F1-u), and DU 165 may be connected to one or more RUs 170 via fronthaul communication links 168 (e.g., an open fronthaul (FH) interface). In some examples, midhaul communication link 162 or fronthaul communication link 168 may be implemented according to interfaces (eg, channels) between layers of a protocol stack supported by respective network entities 105 communicating via those communication links.

[0056] In some wireless communication systems (e.g., wireless communication system 100), infrastructure and spectrum resources for radio access may support wireless backhaul link capabilities to supplement wired backhaul connections, thereby providing an IAB network architecture (e.g., to core network 130). In some cases, in an IAB network, one or more network entities 105 (e.g., IAB nodes 104) may be partially controlled by each other. One or more IAB nodes 104 may be referred to as donor entities or IAB donors. One or more DUs 165 or one or more RUs 170 may be controlled in part by one or more CUs 160 associated with a donor network entity 105 (e.g., donor base station 140). One or more donor network entities 105 (e.g., IAB donors) may communicate with one or more additional network entities 105 (e.g., IAB nodes 104) via supported access and backhaul links (e.g., backhaul communication links 120). The IAB node 104 may include an IAB mobile terminal (IAB-MT) controlled (e.g., scheduled) by a coupled IAB donor's DU 165. The IAB-MT may include an independent set of antennas for relaying communications with the UE 115, or may share the same antennas (e.g., of the RU 170) of the IAB node 104 for access via the DU 165 of the IAB node 104 (e.g., referred to as a virtual IAB-MT (vIAB-MT)). In some examples, the IAB node 104 may include a DU 165 that supports communication links with additional entities (e.g., IAB node 104, UE 115) within a relay chain or configuration (e.g., downstream) of the access network. In such cases, one or more components of the decomposed RAN architecture (e.g., one or more IAB nodes 104 or components of the IAB node 104) may be configured to operate according to the techniques described herein.

[0057] In the case where the techniques described herein are applied to the context of a disaggregated RAN architecture, one or more components of the disaggregated RAN architecture may be configured to support the techniques for reporting correlation metrics for machine learning reproducibility as described herein. For example, some operations described as being performed by a UE 115 or a network entity 105 (e.g., a base station 140) may additionally or alternatively be performed by one or more components of the disaggregated RAN architecture (e.g., an IAB node 104, a DU 165, a CU 160, a RU 170, a RIC 175, a SMO 180).

[0058] UE 115 may include or may be referred to as a mobile device, a wireless device, a remote device, a handheld device, or a subscriber device, or some other suitable term, where a "device" may also be referred to as a unit, a station, a terminal, or a client, etc. UE 115 may also include or may be referred to as a personal electronic device, such as a cellular phone, a personal digital assistant (PDA), a tablet computer, a laptop computer, or a personal computer. In some examples, UE 115 may include or may be referred to as a wireless local loop (WLL) station, an Internet of Things (IoT) device, an Internet of Everything (IoE) device, or a machine type communication (MTC) device, etc., which may be implemented in various objects such as appliances or vehicles, meters, etc.

[0059] The UE 115 described herein may be capable of communicating with various types of devices such as other UEs 115 which may sometimes act as relays, as well as network entities 105 and network equipment including macro eNBs or gNBs, small cell eNBs or gNBs, or relay base stations, etc. Figure 1 shown.

[0060] The UE 115 and the network entity 105 may wirelessly communicate with each other via one or more communication links 125 (e.g., access links) using resources associated with one or more carriers. The term "carrier" may refer to a collection of RF spectrum resources having a physical layer structure defined for supporting the communication link 125. For example, a carrier for the communication link 125 may include a portion of an RF spectrum band (e.g., a bandwidth portion (BWP)) that operates according to one or more physical layer channels for a given radio access technology (e.g., LTE, LTE-A, LTE-A Pro, NR). Each physical layer channel may carry acquisition signaling (e.g., synchronization signals, system information), control signaling for coordinating carrier operations, user data, or other signaling. The wireless communication system 100 may support communications with the UE 115 using carrier aggregation or multi-carrier operation. Depending on the carrier aggregation configuration, the UE 115 may be configured with multiple downlink component carriers and one or more uplink component carriers. Carrier aggregation may be used for both frequency division duplex (FDD) and time division duplex (TDD) component carriers. Communication between the network entity 105 and other devices may refer to communication between these devices and any portion (e.g., entity, sub-entity) of the network entity 105. For example, the terms "send," "receive," or "communicate" when referring to the network entity 105 may refer to any portion of a network entity 105 (e.g., base station 140, CU 160, DU 165, RU 170) of a RAN communicating with another device (e.g., directly or via one or more other network entities 105).

[0061] The signal waveform transmitted via the carrier may include multiple subcarriers (e.g., using multi-carrier modulation (MCM) techniques, such as orthogonal frequency division multiplexing (OFDM) or discrete Fourier transform spread OFDM (DFT-S-OFDM)). In a system using MCM technology, a resource element may refer to a resource of one symbol period (e.g., the duration of one modulation symbol) and one subcarrier, in which case the symbol period and the subcarrier spacing may be inversely related. The number of bits carried by each resource element may depend on the modulation scheme (e.g., the order of the modulation scheme, the coding rate of the modulation scheme, or both), so that a relatively high number of resource elements (e.g., in the transmission duration) and a relatively high order modulation scheme may correspond to a relatively high communication rate. Wireless communication resources may refer to a combination of RF spectrum resources, time resources, and spatial resources (e.g., spatial layers or beams), and the use of multiple spatial resources may increase the data rate or data integrity used for communication with UE 115.

[0062] The time interval for the network entity 105 or the UE 115 may be expressed in multiples of a basic time unit, which may be, for example, a sampling period. seconds, of which can represent the supported subcarrier spacing, and The supported discrete Fourier transform (DFT) size may be indicated. Time intervals of communication resources may be organized according to radio frames, each having a specified duration (e.g., 10 milliseconds (ms)). Each radio frame may be identified by a system frame number (SFN) (e.g., ranging from 0 to 1023).

[0063] Each frame may include a plurality of consecutively numbered subframes or time slots, and each subframe or time slot may have the same duration. In some examples, a frame may be divided into subframes (e.g., in the time domain), and each subframe may be further divided into a certain number of time slots. Alternatively, each frame may include a variable number of time slots, and the number of time slots may depend on the subcarrier spacing. Each time slot may include a certain number of symbol periods (e.g., depending on the length of the cyclic prefix appended to the front of each symbol period). In some wireless communication systems 100, the time slot may be further divided into a plurality of micro-time slots associated with one or more symbols. Excluding the cyclic prefix, each symbol period may be associated with one or more (e.g., The duration of a symbol period may depend on the subcarrier spacing or the operating frequency band.

[0064] A subframe, a time slot, a mini-time slot, or a symbol may be the smallest scheduling unit (e.g., in the time domain) of the wireless communication system 100 and may be referred to as a transmission time interval (TTI). In some examples, the TTI duration (e.g., the number of symbol periods in a TTI) may be variable. Additionally or alternatively, the smallest scheduling unit of the wireless communication system 100 may be dynamically selected (e.g., in a burst of a shortened TTI (sTTI)).

[0065] Physical channels may be multiplexed according to various techniques in order to communicate using a carrier. For example, physical control channels and physical data channels may be multiplexed using one or more of a time division multiplexing (TDM) technique, a frequency division multiplexing (FDM) technique, or a hybrid TDM-FDM technique to signal via a downlink carrier. A control region (e.g., a control resource set (CORESET)) of a physical control channel may be defined by a set of symbol periods and may extend across a system bandwidth of a carrier or a subset of that system bandwidth. One or more control regions (e.g., CORESETs) may be configured for a set of UEs 115. For example, one or more of the UEs 115 may monitor or search the control region for control information according to one or more search space sets, and each search space set may include one or more control channel candidates in one or more aggregation levels arranged in a cascaded manner. The aggregation level of the control channel candidate may refer to the amount of control channel resources (e.g., control channel elements (CCEs)) associated with the coded information for a control information format having a given payload size. The search space sets may include a common search space set configured for transmitting control information to multiple UEs 115 , and a UE-specific search space set for transmitting control information to a specific UE 115 .

[0066] In some examples, the network entities 105 (e.g., base stations 140, RUs 170) may be mobile and thus provide communication coverage for mobile coverage areas 110. In some examples, different coverage areas 110 associated with different technologies may overlap, but the different coverage areas 110 may be supported by the same network entity 105. In some other examples, overlapping coverage areas 110 associated with different technologies may be supported by different network entities 105. The wireless communication system 100 may include, for example, a heterogeneous network in which different types of network entities 105 provide coverage for various coverage areas 110 using the same or different radio access technologies.

[0067] The wireless communication system 100 may be configured to support ultra-reliable communication or low-latency communication or various combinations thereof. For example, the wireless communication system 100 may be configured to support ultra-reliable low-latency communication (URLLC). The UE 115 may be designed to support ultra-reliable or low-latency or critical functions. Ultra-reliable communications may include private communications or group communications and may be supported by one or more services such as push-to-talk, video, or data. Support for ultra-reliable, low-latency functions may include prioritization of services, and such services may be used for public safety or general commercial applications. The terms ultra-reliable, low-latency, and ultra-reliable low-latency may be used interchangeably herein.

[0068] In some examples, a UE 115 may be configured to support communication directly with other UEs 115 via a device-to-device (D2D) communication link 135 (e.g., according to a peer-to-peer (P2P), D2D, or sidelink protocol). In some examples, one or more UEs 115 in a group that are performing D2D communication may be within a coverage area 110 of a network entity 105 (e.g., a base station 140, a RU 170), which may support aspects of such D2D communication configured (e.g., scheduled) by the network entity 105. In some examples, one or more UEs 115 in such a group may be outside of the coverage area 110 of the network entity 105, or may otherwise be unable or not configured to receive transmissions from the network entity 105. In some examples, a group of UEs 115 communicating via D2D communication may support a one-to-many (1:M) system, in which each UE 115 transmits to each of the other UEs 115 in the group. In some examples, network entity 105 may facilitate scheduling of resources for D2D communications. In some other examples, D2D communications may be performed between UEs 115 without involving network entity 105.

[0069] The core network 130 may provide user authentication, access authorization, tracking, Internet Protocol (IP) connectivity, and other access, routing, or mobility functions. The core network 130 may be an evolved packet core (EPC) or a 5G core (5GC), which may include at least one control plane entity (e.g., a mobility management entity (MME), an access and mobility management function (AMF)) for managing access and mobility and at least one user plane entity (e.g., a serving gateway (S-GW), a packet data network (PDN) gateway (P-GW), or a user plane function (UPF)) for routing packets or interconnecting to an external network. The control plane entity may manage non-access stratum (NAS) functions such as mobility, authentication, and bearer management of UEs 115 served by network entities 105 (e.g., base stations 140) associated with the core network 130. User IP packets may be transmitted through the user plane entity, which may provide IP address allocation and other functions. The user plane entity may be connected to the IP services 150 of one or more network operators. IP services 150 may include access to the Internet, an intranet, an IP Multimedia Subsystem (IMS), or packet-switched streaming services.

[0070] The wireless communication system 100 may operate using one or more frequency bands that may be in the range of 300 megahertz (MHz) to 300 gigahertz (GHz). Generally speaking, the region from 300 MHz to 3 GHz is referred to as the ultra-high frequency (UHF) region or decimeter band because the wavelengths range from about one decimeter to one meter in length. UHF waves may be blocked or redirected by buildings and environmental features (which may be referred to as clusters), but these waves may be sufficient to penetrate structures so that macro cells provide service to UEs 115 located indoors. Communications using UHF waves may be associated with smaller antennas and shorter ranges (e.g., less than 100 kilometers) than communications using the lower frequencies and longer wavelengths of the high frequency (HF) or very high frequency (VHF) portions of the spectrum below 300 MHz.

[0071] The wireless communication system 100 can utilize licensed and unlicensed RF spectrum bands. For example, the wireless communication system 100 can use unlicensed bands (such as the 5 GHz Industrial, Scientific, and Medical (ISM) band) to employ License Assisted Access (LAA), LTE Unlicensed (LTE-U) radio access technology, or NR technology. When operating using unlicensed RF spectrum bands, devices such as network entity 105 and UE 115 can employ carrier sensing for conflict detection and avoidance. In some examples, operations performed using unlicensed bands can be based on carrier aggregation configuration (e.g., LAA) in conjunction with component carriers operating using licensed bands. Operations performed using unlicensed spectrum may include downlink transmissions, uplink transmissions, P2P transmissions, or D2D transmissions, among others.

[0072] The network entity 105 (e.g., base station 140, RU 170) or UE 115 may be equipped with multiple antennas that can be used to employ technologies such as transmit diversity, receive diversity, multiple-input multiple-output (MIMO) communication, or beamforming. The antennas of the network entity 105 or UE 115 may be located in one or more antenna arrays or antenna panels, which may support MIMO operations or transmit or receive beamforming. For example, one or more base station antennas or antenna arrays may be co-located at an antenna assembly, such as an antenna tower. In some examples, antennas or antenna arrays associated with the network entity 105 may be located at different geographical locations. The network entity 105 may include an antenna array having a set of antenna ports of multiple rows and columns that the network entity 105 can use to support beamforming for communications with the UE 115. Similarly, the UE 115 may include one or more antenna arrays that may support various MIMO or beamforming operations. Additionally or alternatively, the antenna panel may support RF beamforming for signals sent via the antenna ports.

[0073] Beamforming (which may also be referred to as spatial filtering, directional transmission, or directional reception) is a signal processing technique that may be used at a transmitting device or a receiving device (e.g., a network entity 105, a UE 115) to shape or direct an antenna beam (e.g., a transmit beam, a receive beam) along a spatial path between the transmitting device and the receiving device. Beamforming may be achieved by combining signals communicated via antenna elements of an antenna array so that some signals propagating in a particular direction relative to the antenna array experience constructive interference, while other signals experience destructive interference. Adjustments to signals communicated via antenna elements may include the transmitting device or the receiving device applying an amplitude offset, a phase offset, or both to signals carried via antenna elements associated with the device. Adjustments associated with each of these antenna elements may be defined by a set of beamforming weights associated with a particular direction (e.g., relative to the antenna array of the transmitting device or the receiving device or relative to some other direction).

[0074] The network entity 105 or the UE 115 may use beam scanning techniques as part of a beamforming operation. For example, the network entity 105 (e.g., base station 140, RU 170) may use multiple antennas or antenna arrays (e.g., antenna panels) to perform beamforming operations for directional communication with the UE 115. Some signals (e.g., synchronization signals, reference signals, beam selection signals, or other control signals) may be sent multiple times by the network entity 105 along different directions. For example, the network entity 105 may send signals according to different sets of beamforming weights associated with different transmission directions. Transmissions along different beam directions may be used to identify (e.g., by a transmitting device (such as the network entity 105), or by a receiving device (such as the UE 115)) beam directions for later transmission or reception by the network entity 105.

[0075] Some signals, such as data signals associated with a particular receiving device, may be transmitted by a transmitting device (e.g., transmitting network entity 105, transmitting UE 115) along a single beam direction (e.g., a direction associated with a receiving device (such as receiving network entity 105 or receiving UE 115)). In some examples, a beam direction associated with transmission along a single beam direction may be determined based on signals transmitted along one or more beam directions. For example, UE 115 may receive one or more of the signals sent by network entity 105 along different directions, and may report to network entity 105 an indication of the signal received by UE 115 with the highest signal quality or other acceptable signal quality.

[0076] In some examples, transmission by a device (e.g., by network entity 105 or UE 115) may be performed using multiple beam directions, and the device may use a combination of digital precoding or beamforming to generate a combined beam for transmission (e.g., from network entity 105 to UE 115). UE 115 may report feedback indicating precoding weights for one or more beam directions, and the feedback may correspond to a configured beam set across a system bandwidth or one or more subbands. Network entity 105 may send a reference signal (e.g., a cell-specific reference signal (CRS), a channel state information reference signal (CSI-RS)), which may or may not be precoded. UE 115 may provide feedback for beam selection, which may be a precoding matrix indicator (PMI) or codebook-based feedback (e.g., a multi-panel codebook, a linear combination codebook, a port selection codebook). Although these techniques are described with reference to signals sent along one or more directions by a network entity 105 (e.g., base station 140, RU 170), UE 115 may use similar techniques to send signals multiple times along different directions (e.g., to identify a beam direction for subsequent transmission or reception by UE 115), or to send signals along a single direction (e.g., to send data to a receiving device).

[0077] A receiving device (e.g., UE 115) may perform receiving operations according to multiple receiving configurations (e.g., directional listening) when receiving various signals (such as synchronization signals, reference signals, beam selection signals, or other control signals) from a receiving device (e.g., network entity 105). For example, the receiving device may perform reception according to multiple receiving directions by receiving via different antenna subarrays, processing received signals according to different antenna subarrays, receiving according to different receiving beamforming weight sets (e.g., different directional listening weight sets) applied to signals received at multiple antenna elements of an antenna array, or processing received signals according to different receiving beamforming weight sets applied to signals received at multiple antenna elements of an antenna array, any of which may be referred to as "listening" according to different receiving configurations or receiving directions. In some examples, the receiving device may use a single receiving configuration to receive along a single beam direction (e.g., when receiving a data signal). A single receive configuration may be aligned along a beam direction determined based on listening according to different receive configuration directions (e.g., a beam direction determined to have the highest signal strength, highest signal-to-noise ratio (SNR), or other acceptable signal quality based on listening according to multiple beam directions).

[0078] In some implementations, the wireless communication system 100 may support machine learning techniques to generate predictions (e.g., output functions) based on training data. The training data may be associated with a specific scenario or multiple different scenarios, and may include one or more parameter sets associated with multiple communication environments (e.g., channel environments or coverage areas 110). However, in some cases, the training data may not be a good representation of the test environment (e.g., the location where the test data is obtained by one or more wireless devices) in which the machine learning techniques are implemented. That is, although the training data may be based on one or more communication environments or one or more channel environments or a combination thereof, there may be communication environments or channel environments that are different (e.g., significantly different) from the multiple communication environments associated with the training data. For example, a communication environment (e.g., or channel environment) may have different numbers and positioning of reflectors, clusters, diffraction points, UEs 115, network entities 105, or other objects within the communication environment. Additionally, antenna configurations, antenna arrangements, number of antennas, beamforming capabilities, and beam measurement capabilities may be different for one or more devices (e.g., UEs 115 and network entities 105) within the communication environment.

[0079] According to the examples described herein, a machine learning server may generate a low-dimensional parameter set representing training data. Similarly, one or more devices (e.g., UE 115 and network entity 105) in a communication environment and / or a channel environment may generate a low-dimensional parameter set associated with test data (e.g., data associated with a communication environment or a channel environment). The one or more devices may send a message indicating a low-dimensional parameter set representing test data to a machine learning server. The machine learning server may generate a reproducibility metric based on the correlation between the low-dimensional parameter set representing the training data and the received low-dimensional parameter set representing the test data. The reproducibility metric may be associated with a confidence value of a prediction generated by the machine learning server. The machine learning server may then send a message indicating the reproducibility metric to the one or more devices.

[0080] Thus, the one or more devices may perform communication procedures based on the received reproducibility metric. For example, if the reproducibility metric meets the threshold, the one or more devices may continue to rely on the machine learning technique, or if the reproducibility metric fails to meet the threshold, the one or more devices may perform communication procedures based on measurements performed by the one or more devices.

[0081] Figure 2 An example of a wireless communication system 200 that supports techniques for reporting a correlation metric for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The wireless communication system 200 may include a device 205-a and a device 205-b, which may be as described herein with reference to Figure 1The wireless communication system 200 may also include an example of a communication environment 240, which may be a communication environment as described herein. Figure 1 An example of the coverage area 110 is shown.

[0082] The wireless communication system 200 includes a machine learning server 210. In some examples, the machine learning server 210 can be part of a device 205 (such as a network entity 105, a UE 115, or another device). In other examples, the machine learning server 210 can be a stand-alone device, such as an independent node associated with multiple devices 205 in the network. For example, the machine learning server can be an example of one or more devices 205 in the network (such as a UE 115 or a network entity 105) or can be included in the one or more devices. Additionally or alternatively, the machine learning server 210 can be separate (e.g., a stand-alone device) that communicates with one or more other devices in the network, or is associated with one or more devices in the network, or is common to one or more devices in the network (e.g., provides functionality to the one or more devices), or any combination thereof.

[0083] The machine learning server 210 can communicate with the device 205 via a communication link 245-a, which can be an example of a communication link 125, a backhaul communication link 120, a midhaul communication link 162, or a fronthaul communication link 168, as described herein with reference to Figure 1 Similarly, device 205-a and device 205-b may communicate via communication link 245-b, which may be an example of communication link 125, as described herein with reference to Figure 1 described.

[0084] In some implementations, the wireless communication system 200 may support machine learning techniques to generate predictions (e.g., output functions) based on training data. For example, the machine learning server 210 may be trained (e.g., configured) using training data, and the training data may include one or more parameter sets associated with multiple communication environments 240 (e.g., or channel environments or any combination thereof). For example, the training data may include information about the number and location of clusters 235 (e.g., clusters 235-a, clusters 235-b, which may be examples of blockages, reflectors, diffraction points, etc.), UEs 115, network entities 105, or other objects within multiple communication environments 240 (e.g., or channel environments or combinations thereof). Additionally, the training data may include information about antenna configurations, antenna arrangements, number of antennas, beamforming capabilities, and beam measurement capabilities of one or more devices within the multiple communication environments 240.

[0085] The machine learning server 210 may generate (e.g., predict (e.g., using a machine learning model)) one or more outputs (e.g., parameters, procedures, values, objective functions, or other outputs) in response to input data received from, for example, a communication environment 240 (e.g., a channel environment). The machine learning server 210 may send a message indicating the output to one or more devices within the communication environment 240. For example, the machine learning server 210 may send a message indicating a prediction, which may include a beam weight prediction (e.g., for a blocked scenario or a non-blocked scenario), a predicted reference signal received power (RSRP) value, a predicted rate, a spectral efficiency prediction, a power saving parameter, a blocking mitigation procedure (e.g., a blocking prediction), or other predictions that the one or more devices 205 may use to support performing communications. For example, device 205-a and device 205-b may form or adjust beam 230-a or beam 230-b, respectively, based on the received prediction information. Reference is made to Figure 3 The generation of predictive information and outputs (e.g., using machine learning models) is described in more detail.

[0086] However, in some cases, the training data of the machine learning server 210 may not be a good representation of the communication environment 240 (e.g., or channel environment). For example, although the training data may be based on multiple communication environments 240, the communication environment 240 may be significantly different from the multiple communication environments 240 associated with the training data of the machine learning server 210. For example, the communication environment may have different numbers and positioning of clusters 235 (e.g., which may include reflectors or diffraction points), such as cluster 235-a and cluster 235-b, and devices 205, such as device 205-a and device 205-b, and other objects within the communication environment 240. Additionally or alternatively, the antenna configuration, antenna arrangement, number of antennas, beamforming capabilities, and beam measurement capabilities may be different for one or more of device 205-a and device 205-b. This may cause the prediction output by the machine learning server 210 to be relatively poor when used for communication between device 205-a and device 205-b.

[0087] In some examples, such issues can be assisted by retraining the machine learning server 210 with additional training data that may be relatively larger than the original training information or may represent a larger number of communication environments 240 (e.g., or channel environments, or a combination thereof). However, this may result in a large amount of overhead for the larger amount of training data. Additionally, communicating the larger amount of training data to the machine learning server 210 may be associated with a large amount of latency and may result in overtraining of the machine learning server 210 and a poor fit of the machine learning model associated with the machine learning server 210.

[0088] Therefore, according to the examples disclosed herein, the machine learning server 210 may generate a low-dimensional parameter set representing the training data. For example, the machine learning server 210 may reduce the dimensionality of the training data by abstracting the training data (e.g., the data may be 100-dimensional) that may contain a relatively large number of dimensions into a low-dimensional parameter set (e.g., the parameter set may be 8-dimensional) that may contain a relatively low number of dimensions. Similarly, the device 205-a and the device 205-b may generate a low-dimensional parameter set representing test data, which may be data associated with the communication environment 240. In some examples, the device 205-a, the device 205-b, the machine learning server 210, or a combination thereof may reduce the dimensionality of the training data or the test data based at least in part on various dimensionality reduction techniques such as principal component analysis, non-negative matrix decomposition, linear discriminant analysis, generalized discriminant analysis, and the like. In this case, the data in the high-dimensional space (e.g., training data, test data) may be transformed into data in the low-dimensional space that represents the high-dimensional data (e.g., retaining certain characteristics and / or attributes of the high-dimensional data).

[0089] The test data may include parameters collected by the device 205-a and / or the device 205-b, or information about the device 205-a and / or the device 205-b. For example, the test data may include information about the number and positioning of the clusters 235-a and the clusters 235-b, the locations of the devices 205-a and the devices 205-b, or the locations of other objects within the communication environment 240. Additionally or alternatively, the test data may include information about the antenna configuration, antenna arrangement, number of antennas, beamforming capabilities, and beam measurement capabilities of the devices 205-a and the devices 205-b, and information about the beams 230-a and / or the beams 230-b associated with the devices 205-a and the devices 205-b, respectively.

[0090] In some examples, the machine learning server 210, the device 205, or both may use principal component analysis (and other techniques) to abstract the training data or test data into a low-dimensional parameter set representing the training data or test data, respectively. As an illustrative example, principal component analysis can be used to reduce the dimension of one or more relatively large data sets, for example, by transforming a set of variables into a relatively small set of variables, which relatively small set of variables includes most of the information in a larger set. Here, the principal component used for principal component analysis may refer to a variable constructed as a linear combination of the (initial) variables in the original (e.g., relatively large) data set. Such a combination is generated in the following manner: the variables (e.g., principal components) may be unrelated, and a certain amount (e.g., most) of the information within the initial variables is compressed into the principal components. In other words, principal component reduction can reduce the dimension of the original data set while minimizing information loss, which may include discarding some components (e.g., with relatively low information) and treating the remaining components as variables (e.g., principal components) in the new data set. The machine learning server 210 or device 205 may accordingly determine one or more parameters (eg, dimensions) of the training data or the test data that may be used to closely represent the entirety of the training data or the test data, respectively.

[0091] In some cases, the device 205-a and the device 205-b may collaborate to generate a low-dimensional parameter set representing the test data. For example, the device 205-b may transmit a message to the device 205-a containing a parameter indication 215 indicating one or more parameters, which may be information associated with the device 205-b or information about the communication environment 240 obtained by the device 205-b. The device 205-a may then generate a low-dimensional parameter set representing the test data, which includes the one or more parameters received from the device 205-b, other parameters known (e.g., obtained or measured) by the device 205-a, parameters received from other devices 205, or a combination thereof.

[0092] Device 205-a or device 205-b or both may send a message including parameter set indication 220, which may indicate to machine learning server 210 a low-dimensional parameter set representing test data associated with communication environment 240. In some cases, such as when device 205-a and device 205-b collaborate to generate a low-dimensional parameter set, device 205-b may avoid sending a message including parameter set indication 220 to machine learning server 210.

[0093] The machine learning server 210 may thus receive one or more messages from one or more devices 205 including a parameter set indication 220 indicating a low-dimensional parameter set associated with the communication environment 240. In some cases, such as when the machine learning server 210 receives multiple parameter set indications 220, the machine learning server 210 may combine the multiple low-dimensional parameter sets received from the multiple devices 205 into a single low-dimensional parameter set representing the test data for the communication environment 240.

[0094] The machine learning server 210 may associate a low-dimensional parameter set representing training data for the machine learning server 210 with a low-dimensional parameter set representing test data associated with the communication environment 240. In some examples, the machine learning server 210 may associate the low-dimensional parameter set representing the training data with the low-dimensional parameter set representing the test data by associating principal components associated with each low-dimensional data set in the low-dimensional parameter set. For example, each low-dimensional parameter set may include one or more principal components (e.g., if the machine learning server 210, the device 205, or a combination thereof performed a principal component analysis to obtain the low-dimensional parameter set), and the machine learning server 210 may associate the one or more principal components in each of the low-dimensional parameter sets. In some examples, the principal components may be one or more unit vectors fitted to the training data or the test data.

[0095] The machine learning server 210 may generate a reproducibility metric that is or represents the correlation performed by the machine learning server 210. For example, the reproducibility metric may include a value or other indication of how well the test data correlates with the training data. Additionally or alternatively, the reproducibility metric may be or represent a confidence value (e.g., a score) associated with the prediction made by the machine learning server 210 regarding the communication environment 240.

[0096] In some cases, if the machine learning server 210 determines that the reproducibility metric does not meet the threshold, the machine learning server 210 may trigger a retraining procedure. For example, the machine learning server 210 may generate (e.g., request, obtain, or compile) new training data, which in some cases may include parameters associated with the communication environment 240. The machine learning server 210 may then retrain using the new training data so that predictions from the machine learning model generated by the machine learning server 210 can be used by the device 205-a and the device 205-b with a higher degree of accuracy.

[0097] The machine learning server 210 may send a message including a reproducibility metric indication 225-a to the device 205-a and the device 205-b, which may indicate a reproducibility metric generated by the machine learning server 210. Additionally or alternatively, the device 205-a and the device 205-b may collaborate to share information about the reproducibility metric, as illustrated. For example, the machine learning server 210 may send a message including a reproducibility metric indication 225-a to the device 205-a. The device 205-a may then send a message including a reproducibility metric indication 225-b to the device 205-b to indicate the reproducibility metric to the device 205-b.

[0098] Device 205-a and device 205-b may perform communication procedures based on the reproducibility metric. For example, device 205-a and device 205-b may decide whether to use predictions associated with machine learning server 210 in the communication procedures. For example, if the reproducibility metric meets the threshold, device 205-a and device 205-b may continue to rely on machine learning techniques. Alternatively, if the reproducibility metric fails to meet the threshold, device 205-a and device 205-b may perform communication procedures based on one or more measurements performed by device 205-a and device 205-b.

[0099] Thus, if the test data associated with the communication environment 240 does not correlate well with the training data of the machine learning server 210, the device 205-a and the device 205-b may avoid using predictions from the machine learning model generated by the machine learning server 210. In other cases, when the test data associated with the communication environment 240 does correlate with the training data of the machine learning server 210 with some degree of accuracy (e.g., the test data may be "close enough" to use based on the training data), the predictions from the machine learning model generated by the machine learning server 210 may be used. In any case, the reproducibility metrics may enable improvements to the communication procedures at the device 205-a and the device 205-b, and may result in improved communication stability and user experience.

[0100] Figure 3 An example of a machine learning process 300 that supports techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated. The machine learning process 300 may be executed on a machine learning server (such as the one referenced herein). Figure 2 The machine learning process 300 may support training a machine learning model (e.g., an artificial neural network) for network power saving, UE power saving, load balancing, mobility management, beamforming, RSRP, rate, or spectrum efficiency prediction, or any combination of these or other processes improved by machine learning, as described herein with reference to Figure 1 and Figure 2 described.

[0101] The machine learning process 300 may include a machine learning algorithm 310. As shown, the machine learning algorithm 310 may be an example of a neural network (e.g., an artificial neural network), such as a FF or DFF neural network, an RNN, an LSTM neural network, or any other type of neural network. However, any other machine learning algorithm may be supported. For example, the machine learning algorithm 310 may implement a nearest neighbor algorithm, a linear regression algorithm, a naive Bayes algorithm, a random forest algorithm, or any other machine learning algorithm. In addition, the machine learning process 300 may involve supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, or any combination thereof.

[0102] The machine learning algorithm 310 may include an input layer 315, one or more hidden layers 320, and an output layer 325. In a fully connected neural network with one hidden layer 320, each hidden layer node 335 (e.g., hidden layer node 335-a to hidden layer node 335-d) may receive values ​​as input from each input layer node 330 (e.g., input layer node 330-a to input layer node 330-c), where each input may be weighted. These neural network weights may be based on a cost function modified during training of the machine learning algorithm 310. Similarly, each output layer node 340 (e.g., output layer node 340-a to output layer node 340-c) may receive values ​​as input from each hidden layer node 335, where these inputs are weighted. If post-deployment training (e.g., online training) is supported, memory may be allocated to store errors or gradients of inverse matrix multiplications. These errors or gradients may support updating the machine learning algorithm 310 based on output feedback. Training the machine learning algorithm 310 may support the calculation of weights (e.g., connecting the input layer nodes 330 to the hidden layer nodes 335 and connecting the hidden layer nodes 335 to the output layer nodes 340) to map the input pattern to the desired output result. The training may generate a device-specific machine learning algorithm 310 based on historical application data and data transmission of a specific network entity 105 or UE 115.

[0103] In some examples, input values ​​305 may be transmitted to machine learning algorithm 310 for processing. In some examples, pre-processing may be performed according to a sequence of operations on input values ​​305 so that input values ​​305 may be in a format compatible with machine learning algorithm 310. Input values ​​305 may be converted at input layer 315 into 305. In some cases, different measurements may be input at different input layer nodes 330 of the input layer 315. If the number of input layer nodes 330 exceeds the number of inputs corresponding to the input values ​​305, some of the input layer nodes 330 may be assigned default values ​​(e.g., a value of 0). As shown, the input layer 315 may include three input layer nodes 330-a, 330-b, and 330-c. However, it should be understood that the input layer 315 may include any number of input layer nodes 330 (e.g., 20 input nodes).

[0104] The machine learning algorithm 310 may be based on The input layer nodes 330 and The number of input to hidden weights between hidden layer nodes 335 is used to convert the input layer 315 into a hidden layer 320. The machine learning algorithm 310 may include any number of hidden layers 320 as an intermediate step between the input layer 315 and the output layer 325. Additionally, each hidden layer 320 may include any number of nodes. For example, as shown in the figure, the hidden layer 320 may include four hidden layer nodes 335-a, 335-b, 335-c and 335-d. However, it should be understood that the hidden layer 320 may include any number of hidden layer nodes 335 (e.g., 10 input nodes). In a fully connected neural network, each node in a layer may be based on each node in a previous layer. For example, the value of a hidden layer node 335-a may be based on the values ​​of input layer nodes 330-a, 330-b and 330-c (e.g., different weights are applied to each node value).

[0105] The machine learning algorithm 310 may determine the value of the output layer node 340 of the output layer 325 after the one or more hidden layers 320. For example, the machine learning algorithm 310 may be based on There are 335 hidden layer nodes and The number of hidden to output weights between the output layer nodes 340 transforms the hidden layer 320 into the output layer 325. In some cases, . Each output layer node 340 may correspond to a different output value 345 of the machine learning algorithm 310. As shown, the machine learning algorithm 310 may include three output layer nodes 340-a, 340-b, and 340-c, which support three different thresholds. However, it should be understood that the output layer 325 may include any number of output layer nodes 340. In some examples, post-processing may be performed on the output value 345 according to the sequence of operations so that the output value 345 may be in a format compatible with the report output value 345.

[0106] The machine learning process 300 may support model training for network power conservation, UE power conservation, load balancing, mobility management, channel measurement, beam management, or any other functionality supported by the network entity 105 or UE 115. For example, a device (e.g., a network entity 105, a UE 115, or other training device) may train a machine learning algorithm 310 (e.g., a machine learning model, a neural network) using one or more security techniques to mitigate negative impacts from corrupted data. In some examples, the device may receive one or more indications (e.g., one or more trust scores) of whether the data is trustworthy. If the device determines that a subset of the data is untrustworthy, the device may avoid using the untrustworthy data to train the machine learning algorithm 310. Instead, the device (e.g., a network entity 105) may use trusted data (e.g., from one or more trusted UEs 115) to train the machine learning algorithm 310.

[0107] For example, the machine learning algorithm 310 may be trained using training data (e.g., trusted data) to receive a set of input values ​​305, which may represent cluster data 350, device data 355, antenna parameters 360, beam parameters 365, or any combination of these or other input parameters of the machine learning algorithm 310. The machine learning algorithm 310 may process the set of input values ​​305, and based on the processing, may output a set of output values ​​345, which may represent one or more output predictions 370 or other output parameters of the machine learning algorithm 310. For example, the output predictions 370 may be used by a device (e.g., a network entity 105, a UE 115, or another device) to perform communication procedures, such as for adjusting beams, predicting blocking or clustering, adjusting antenna parameters, or any other operation. Additionally or alternatively, the machine learning algorithm 310 may support other input values ​​305, output values ​​345, or both to support other AI-based improvements to the wireless communication system.

[0108] In some examples, a machine learning server associated with the machine learning algorithm 310 may generate a low-dimensional parameter set (e.g., a low-dimensional data set) representing training data. The machine learning server may also receive a low-dimensional parameter set (e.g., a low-dimensional data set) representing test data associated with the communication environment or the channel environment from one or more devices (e.g., UE 115 and network entity 105) within the communication environment (e.g., or channel environment). The machine learning server may generate a reproducibility metric based on the correlation between the low-dimensional parameter set representing the training data and the received low-dimensional parameter set representing the test data. The reproducibility metric may be associated with a confidence value associated with the prediction generated by the machine learning server. The machine learning server may then send a message indicating the reproducibility metric to the one or more devices.

[0109] In some examples, such as if the machine learning server determines that the reproducibility metric does not meet a threshold, the machine learning server can trigger a retraining procedure. For example, the machine learning server can generate (e.g., request, obtain, or compile) new training data, which in some cases can include parameters associated with the communication environment. The machine learning server can then train (e.g., retrain) using the new training data so that predictions from the machine learning algorithm 310 generated by the machine learning server can be used by the one or more devices with a higher degree of accuracy.

[0110] The one or more devices may perform a communication procedure based on the reproducibility metric received from the machine learning server. For example, the one or more devices may decide whether to use the prediction from the machine learning algorithm 310 in the communication procedure. For example, if the reproducibility metric meets the threshold, the one or more devices may continue to rely on the machine learning technique. Alternatively, if the reproducibility metric fails to meet the threshold, the one or more devices may perform a communication procedure based on one or more measurements performed by the one or more devices.

[0111] Thus, the one or more devices may use or refrain from using predictions 370 from the machine learning algorithm 310 generated by the machine learning server based on whether the test data associated with the communication and / or channel environment correlates with the training data of the machine learning server. Such correlations may be provided to various devices in the system via reproducibility metrics signaled by the machine learning server, which may improve communication procedures at the one or more devices and may result in improved communication stability and user experience.

[0112] Figure 4 An example of a process flow 400 supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated. The process flow 400 may be found in the referenced Figure 1 and Figure 2 The process flow 400 may include a device 405, which may be a UE, a network entity, or a wireless communication system 100 or 200 as described herein. Figures 1 to 3 Another example of a device described herein. Process flow 400 includes a machine learning server 410, which can be a machine learning server as described herein with reference to Figure 2An example of the machine learning server 210 described. For example, the machine learning server 410 can be part of or included in another device such as a network entity or UE. In other examples, the machine learning server 410 can be part of or included in an independent (e.g., self-contained) device that is common to one or more other devices (such as one or more network nodes) and communicates with the one or more other devices. In some aspects, the machine learning server 410 may be associated with one or more functions or nodes of the network. In the following description of the process flow 400, the operations performed by the device may be performed in a different order or at different times. Additionally or alternatively, some operations may be omitted from the process flow 400 and other operations may be added to the process flow 400.

[0113] At 415, the machine learning server 410 may generate a first set of parameters associated with training data, the training data corresponding to a plurality of communication environments (e.g., or channel environments, or a combination thereof). The first set of parameters may include a low-dimensional parameter set representing the training data. Similarly, the device 405 may generate a second set of parameters associated with test data, the test data corresponding to a communication environment (e.g., or channel environment). The second set of parameters may include a low-dimensional parameter set representing the test data. In some examples, the machine learning server 410, the device 405, or both may use principal component analysis to generate a parameter set by abstracting the training data or the test data into a low-dimensional parameter set representing the training data or the test data, respectively. This article combines Figure 2 The generation of low-dimensional parameter sets is described in more detail.

[0114] At 420, device 405 may send a first message indicating a second set of parameters associated with the test data to the machine learning server. In some examples, machine learning server 410 may receive additional messages indicating parameter sets from other devices within the communication environment (e.g., or channel environment). In these examples, machine learning server 410 may combine the parameter sets into a single parameter set associated with the communication environment (e.g., or channel environment).

[0115] At 425, the machine learning server 410 may generate a reproducibility metric. The reproducibility metric may be based on a correlation between the first set of parameters and the second set of parameters. In some examples, the machine learning server 410 may associate the first set of parameters with the second set of parameters by associating principal components associated with each of the first set of parameters and the second set of parameters. The reproducibility metric may be or represent a correlation performed by the machine learning server 410. For example, the reproducibility metric may represent the degree of correlation between the test data and the training data. Additionally or alternatively, the reproducibility metric may be or indicate a confidence value (e.g., a score) associated with the prediction of the machine learning server 410 regarding the communication environment (e.g., or channel environment).

[0116] At 430, machine learning server 410 may send a second message indicating the reproducibility metric to device 405. In some examples, machine learning server 410 may send additional messages indicating the reproducibility metric to one or more other devices within the communication environment (eg, or channel environment).

[0117] At 435, device 405 may perform a communication procedure based on the reproducibility metric. In some cases, device 405 may determine that the reproducibility metric satisfies a threshold, and device 405 may perform one or more communication procedures based on one or more predictions made by machine learning server 410 using training data. In other cases, device 405 may determine that the reproducibility metric fails to satisfy the threshold, and device 405 may perform one or more communication procedures based on one or more measurements performed by device 405. In some examples, device 405 may send a third message indicating the reproducibility metric to another device within the communication environment (e.g., or channel environment).

[0118] Figure 5 A block diagram 500 of a device 505 supporting techniques for reporting relevance metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The device 505 may be an example of aspects of a machine learning server 210, a UE 115, or a network entity 105 as described herein. The device 505 may include a receiver 510, a transmitter 515, and a communication manager 520. The device 505 may also include a processor. Each of these components may communicate with each other (e.g., via one or more buses).

[0119] Receiver 510 may provide means for receiving information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to techniques for reporting correlation metrics for machine learning reproducibility). The information may be communicated to other components of device 505. Receiver 510 may utilize a single antenna or a collection of multiple antennas.

[0120] Transmitter 515 may provide means for transmitting signals generated by other components of device 505. For example, transmitter 515 may transmit information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to techniques for reporting correlation metrics for machine learning reproducibility). In some examples, transmitter 515 may be co-located with receiver 510 in a transceiver module. Transmitter 515 may utilize a single antenna or a collection of multiple antennas.

[0121] The communication manager 520, the receiver 510, the transmitter 515, or various combinations thereof or various components thereof may be examples of means for performing various aspects of the techniques for reporting relevance metrics for machine learning reproducibility as described herein. For example, the communication manager 520, the receiver 510, the transmitter 515, or various combinations thereof or components thereof may support methods for performing one or more of the functions described herein.

[0122] In some examples, the communication manager 520, the receiver 510, the transmitter 515, or various combinations or components thereof may be implemented in hardware (e.g., in a communication management circuit). The hardware may include a processor, a digital signal processor (DSP), a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a microcontroller, discrete gate or transistor logic components, discrete hardware components, or any combination thereof that is configured as or otherwise supports components for performing the functions described in the present disclosure. In some examples, the processor and a memory coupled to the processor may be configured to perform one or more of the functions described herein (e.g., by executing instructions stored in the memory by the processor).

[0123] Additionally or alternatively, in some examples, the communication manager 520, the receiver 510, the transmitter 515, or various combinations or components thereof may be implemented in code executed by a processor (e.g., as communication management software or firmware). If implemented in code executed by a processor, the functionality of the communication manager 520, the receiver 510, the transmitter 515, or various combinations or components thereof may be performed by a general purpose processor (e.g., configured as or otherwise supporting components for performing the functionality described in the present disclosure), a DSP, a CPU, an ASIC, an FPGA, a microcontroller, or any combination of these or other programmable logic devices.

[0124] In some examples, communication manager 520 may be configured to perform various operations (e.g., receive, obtain, monitor, output, transmit) using or otherwise cooperating with receiver 510, transmitter 515, or both. For example, communication manager 520 may receive information from receiver 510, transmit information to transmitter 515, or be integrated in conjunction with receiver 510, transmitter 515, or both to obtain information, output information, or perform various other operations as described herein.

[0125] According to examples as disclosed herein, the communication manager 520 may support wireless communications at a machine learning server. For example, the communication manager 520 may be configured to or otherwise support a component for generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data. The communication manager 520 may be configured to or otherwise support a component for receiving one or more messages from one or more devices, the one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The communication manager 520 may be configured to or otherwise support a component for sending a message to the one or more devices, the message indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0126] Additionally or alternatively, according to examples as disclosed herein, the communication manager 520 may support wireless communication at the first device. For example, the communication manager 520 may be configured to or otherwise support components for generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The communication manager 520 may be configured to or otherwise support components for sending a first message to a machine learning server, the first message indicating the second set of parameters representing the test data. The communication manager 520 may be configured to or otherwise support components for receiving a second message indicating a reproducibility metric based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0127] By including or configuring a communication manager 520 according to examples as described herein, a device 505 (e.g., a processor controlling or otherwise coupled to a receiver 510, a transmitter 515, a communication manager 520, or a combination thereof) may support techniques for reporting relevance metrics for machine learning reproducibility that may provide improved communication reliability, an improved user experience associated with reduced processing, and more efficient utilization of communication resources.

[0128] Figure 6 A block diagram 600 of a device 605 supporting techniques for reporting relevance metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The device 605 may be an example of aspects of the device 505, machine learning server 210, UE 115, or network entity 105 as described herein. The device 605 may include a receiver 610, a transmitter 615, and a communication manager 620. The device 605 may also include a processor. Each of these components may communicate with each other (e.g., via one or more buses).

[0129] Receiver 610 may provide means for receiving information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to techniques for reporting correlation metrics for machine learning reproducibility). The information may be communicated to other components of device 605. Receiver 610 may utilize a single antenna or a collection of multiple antennas.

[0130] Transmitter 615 may provide means for transmitting signals generated by other components of device 605. For example, transmitter 615 may transmit information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to techniques for reporting correlation metrics for machine learning reproducibility). In some examples, transmitter 615 may be co-located with receiver 610 in a transceiver module. Transmitter 615 may utilize a single antenna or a collection of multiple antennas.

[0131] Device 605 or its various components may be examples of components for performing various aspects of the technology for reporting relevance metrics for machine learning reproducibility as described herein. For example, communication manager 620 may include training parameter component 625, test parameter manager 630, reproducibility metric component 635, test parameter component 640, reproducibility metric manager 645, or any combination thereof. Communication manager 620 may be an example of various aspects of communication manager 520 as described herein. In some examples, communication manager 620 or its various components may be configured to perform various operations (e.g., receive, obtain, monitor, output, send) using or otherwise cooperating with receiver 610, transmitter 615, or both. For example, communication manager 620 may receive information from receiver 610, transmit information to transmitter 615, or integrate with receiver 610, transmitter 615, or both to obtain information, output information, or perform various other operations as described herein.

[0132] According to an example as disclosed herein, the communication manager 620 may support wireless communication at a machine learning server. The training parameter component 625 may be configured to or otherwise support a component for generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data. The test parameter manager 630 may be configured to or otherwise support a component for receiving one or more messages from one or more devices, the one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The reproducibility metric component 635 may be configured to or otherwise support a component for sending a message to the one or more devices, the message indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0133] Additionally or alternatively, according to examples as disclosed herein, the communication manager 620 may support wireless communication at the first device. The test parameter component 640 may be configured to or otherwise support components for generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The test parameter manager 630 may be configured to or otherwise support components for sending a first message to a machine learning server, the first message indicating the second set of parameters representing the test data. The reproducibility metric manager 645 may be configured to or otherwise support components for receiving a second message indicating a reproducibility metric based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0134] Figure 7 A block diagram 700 of a communication manager 720 supporting techniques for reporting relevance metrics for machine learning reproducibility in accordance with one or more aspects of the present disclosure is illustrated. The communication manager 720 may be an example of aspects of the communication manager 520, the communication manager 620, or both as described herein. The communication manager 720 or its various components may be examples of components for performing various aspects of the techniques for reporting relevance metrics for machine learning reproducibility as described herein. For example, the communication manager 720 may include a training parameter component 725, a test parameter manager 730, a reproducibility metric component 735, a test parameter component 740, a reproducibility metric manager 745, a retraining component 750, a collaboration component 755, a principal component manager 760, or any combination thereof. Each of these components may communicate directly or indirectly with each other (e.g., via one or more buses), which communication may include communication within a protocol layer of a protocol stack, communication associated with a logical channel of the protocol stack (e.g., between protocol layers of a protocol stack, within a device, component, or virtualized component associated with the network entity 105, between devices, components, or virtualized components associated with the network entity 105), or any combination thereof.

[0135] According to an example as disclosed herein, the communication manager 720 may support wireless communication at a machine learning server. The training parameter component 725 may be configured to or otherwise support a component for generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data. The test parameter manager 730 may be configured to or otherwise support a component for receiving one or more messages from one or more devices, the one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The reproducibility metric component 735 may be configured to or otherwise support a component for sending a message to the one or more devices, the message indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0136] In some examples, the reproducibility metric instructs the one or more devices to perform one or more communication procedures based on one or more predictions from a machine learning model generated by a machine learning server using training data, and the reproducibility metric instructs to perform the one or more communication procedures based on the reproducibility metric satisfying a threshold.

[0137] In some examples, the reproducibility metric indicates to the one or more devices to perform one or more communication procedures based on one or more measurements performed by the one or more devices, the reproducibility metric indicating to perform the one or more communication procedures based on the reproducibility metric failing to meet a threshold.

[0138] In some examples, the reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by a machine learning server using training data applied to the communication environment.

[0139] In some examples, predictions from the machine learning model generated by the machine learning server include beam weight predictions for non-blocking scenarios or blocking scenarios.

[0140] In some examples, reproducibility metric component 735 can be configured or otherwise support components for generating a reproducibility metric based on a correlation between a first set of parameters and a second set of parameters.

[0141] In some examples, to support generating a reproducibility metric, principal component manager 760 may be configured to or otherwise support a component for generating a reproducibility metric based on a correlation between one or more principal components associated with training data and one or more principal components associated with test data, wherein a first set of parameters includes the one or more principal components associated with the training data, and wherein a second set of parameters includes the one or more principal components associated with the test data.

[0142] In some examples, retraining component 750 can be configured to or otherwise support components for generating additional training data for the one or more communication environments or the one or more channel environments, or a combination thereof, based on the reproducibility metric failing to meet a threshold.

[0143] In some examples, the training data includes an indication of at least one of: the number of clusters in the one or more communication environments or the one or more channel environments or a combination thereof, the number of reflectors in the one or more communication environments or the one or more channel environments or a combination thereof, the number of diffraction points in the one or more communication environments or the one or more channel environments or a combination thereof, positioning information associated with one or more devices in the one or more communication environments or the one or more channel environments or a combination thereof, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

[0144] In some examples, the test data includes an indication of at least one of: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the one or more devices, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

[0145] In some examples, the machine learning server is part of the UE or part of a network entity, or comprises a standalone node associated with a collection of multiple devices in the network.

[0146] Additionally or alternatively, according to examples as disclosed herein, the communication manager 720 may support wireless communication at the first device. The test parameter component 740 may be configured to or otherwise support components for generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. In some examples, the test parameter manager 730 may be configured to or otherwise support components for sending a first message to a machine learning server, the first message indicating the second set of parameters representing the test data. The reproducibility metric manager 745 may be configured to or otherwise support components for receiving a second message indicating a reproducibility metric based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0147] In some examples, the reproducibility metric indicates that one or more communication procedures are to be performed according to one or more predictions from a machine learning model generated by a machine learning server using training data, and the reproducibility metric indicates that the first device performs the one or more communication procedures based on the reproducibility metric satisfying a threshold.

[0148] In some examples, the reproducibility metric indicates that one or more communication procedures are performed according to one or more measurements performed by the first device, the reproducibility metric indicating that the first device performs the one or more communication procedures based on the reproducibility metric failing to meet a threshold.

[0149] In some examples, the reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by a machine learning server using training data applied to a communication environment or a channel environment, or both.

[0150] In some examples, predictions from a machine learning model associated with training data generated by a machine learning server include beam weight predictions for a non-blocking scenario or a blocking scenario.

[0151] In some examples, collaboration component 755 may be configured or otherwise support components for receiving a message indicating one or more parameters from a second device within the communication environment or the channel environment or both, wherein generating the second set of parameters is based on receiving the message indicating the one or more parameters.

[0152] In some examples, collaboration component 755 can be configured or otherwise support means for sending a message indicating the reproducibility metric to a second device within the communication environment or the channel environment or both based on receiving the reproducibility metric.

[0153] In some examples, the test data includes an indication of at least one of: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the first device, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the first device, or the beam measurement capability of the first device.

[0154] In some examples, the training data includes an indication of at least one of: the number of clusters in the one or more communication environments or the one or more channel environments or a combination thereof, the number of reflectors in the one or more communication environments or the one or more channel environments or a combination thereof, the number of diffraction points in the one or more communication environments or the one or more channel environments or a combination thereof, positioning information associated with one or more devices in the one or more communication environments or the one or more channel environments or a combination thereof, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

[0155] Figure 8 A diagram of a system 800 including a device 805 that supports techniques for reporting relevance metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The device 805 may be an example of a device 505, a device 605, a machine learning server 210, or a UE 115 as described herein or include components thereof. The device 805 may communicate (e.g., wirelessly) with one or more network entities 105, one or more UEs 115, or any combination thereof. The device 805 may include components for two-way voice and data communications, including components for sending and receiving communications, such as a communication manager 820, an input / output (I / O) controller 810, a transceiver 815, an antenna 825, a memory 830, a code 835, and a processor 840. These components may be in electronic communication or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more buses (e.g., a bus 845).

[0156] I / O controller 810 can manage input signals and output signals of device 805. I / O controller 810 can also manage peripheral devices that are not integrated into device 805. In some cases, I / O controller 810 can represent a physical connection or port to an external peripheral device. In some cases, I / O controller 810 can utilize an operating system, such as iOS. ® ANDROID ® , MS-DOS ®、MS-WINDOWS ® , OS / 2 ® ,UNIX ® 、LINUX ® or another known operating system. Additionally or alternatively, I / O controller 810 may represent or interact with a modem, keyboard, mouse, touch screen, or similar device. In some cases, I / O controller 810 may be implemented as part of a processor (such as processor 840). In some cases, a user may interact with device 805 via I / O controller 810 or via hardware components controlled by I / O controller 810.

[0157] In some cases, the device 805 may include a single antenna 825. However, in some other cases, the device 805 may have more than one antenna 825, which may be capable of concurrently transmitting or receiving multiple wireless transmissions. The transceiver 815 may communicate bidirectionally via one or more antennas 825, a wired or wireless link, as described herein. For example, the transceiver 815 may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. The transceiver 815 may also include a modem for: modulating packets; providing the modulated packets to one or more antennas 825 for transmission; and demodulating packets received from one or more antennas 825. The transceiver 815 or the transceiver 815 and one or more antennas 825 may be examples of transmitters 515, transmitters 615, receivers 510, receivers 610, or any combination thereof or components thereof as described herein.

[0158] The memory 830 may include random access memory (RAM) and read-only memory (ROM). The memory 830 may store computer-readable, computer-executable code 835 including instructions that, when executed by the processor 840, cause the device 805 to perform various functions described herein. The code 835 may be stored in a non-transitory computer-readable medium such as a system memory or another type of memory. In some cases, the code 835 may not be directly executable by the processor 840, but may (for example, when compiled and executed) cause the computer to perform the functions described herein. In some cases, the memory 830 may also contain, among other things, a basic I / O system (BIOS) that may control basic hardware or software operations, such as interaction with peripheral components or devices.

[0159] The processor 840 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, a discrete gate or transistor logic component, a discrete hardware component, or any combination thereof). In some cases, the processor 840 may be configured to operate a memory array using a memory controller. In some other cases, the memory controller may be integrated into the processor 840. The processor 840 may be configured to execute computer-readable instructions stored in a memory (e.g., a memory 830) to cause the device 805 to perform various functions (e.g., functions or tasks supporting techniques for reporting correlation metrics for machine learning reproducibility). For example, the device 805 or a component of the device 805 may include a processor 840 and a memory 830 coupled to or coupled to the processor 840, and the processor 840 and the memory 830 are configured to perform the various functions described herein.

[0160] According to examples as disclosed herein, the communication manager 820 may support wireless communications at a machine learning server. For example, the communication manager 820 may be configured to or otherwise support a component for generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data. The communication manager 820 may be configured to or otherwise support a component for receiving one or more messages from one or more devices, the one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The communication manager 820 may be configured to or otherwise support a component for sending a message to the one or more devices, the message indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0161] Additionally or alternatively, according to examples as disclosed herein, the communication manager 820 may support wireless communication at the first device. For example, the communication manager 820 may be configured to or otherwise support components for generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The communication manager 820 may be configured to or otherwise support components for sending a first message to a machine learning server, the first message indicating the second set of parameters representing the test data. The communication manager 820 may be configured to or otherwise support components for receiving a second message indicating a reproducibility measure based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0162] By including or configuring a communication manager 820 according to examples as described herein, the device 805 may support techniques for reporting relevance metrics for machine learning reproducibility that may provide improved communication reliability, an improved user experience associated with reduced processing, and more efficient utilization of communication resources.

[0163] In some examples, the communication manager 820 may be configured to perform various operations (e.g., receive, monitor, transmit) using or otherwise cooperating with the transceiver 815, one or more antennas 825, or any combination thereof. Although the communication manager 820 is illustrated as a separate component, in some examples, one or more functions described with reference to the communication manager 820 may be supported or performed by the processor 840, the memory 830, the code 835, or any combination thereof. For example, the code 835 may include instructions executable by the processor 840 to cause the device 805 to perform various aspects of the techniques for reporting relevance metrics for machine learning reproducibility as described herein, or the processor 840 and the memory 830 may be otherwise configured to perform or support such operations.

[0164] Fig. 9A diagram of a system 900 including a device 905 supporting techniques for reporting correlation metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The device 905 may be an example of a device 505, a device 605, a machine learning server 210, or a network entity 105 as described herein or include components thereof. The device 905 may communicate with one or more network entities 105, one or more UEs 115, or any combination thereof, which may include communication through one or more wired interfaces, through one or more wireless interfaces, or any combination thereof. The device 905 may include components that support output and acquisition of communications, such as a communication manager 920, a transceiver 910, an antenna 915, a memory 925, a code 930, and a processor 935. These components may be in electronic communication or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more buses (e.g., bus 940).

[0165] The transceiver 910 may support bidirectional communication via a wired link, a wireless link, or both as described herein. In some examples, the transceiver 910 may include a wired transceiver and may communicate bidirectionally with another wired transceiver. Additionally or alternatively, in some examples, the transceiver 910 may include a wireless transceiver and may communicate bidirectionally with another wireless transceiver. In some examples, the device 905 may include one or more antennas 915, which may be capable of sending or receiving wireless transmissions (e.g., concurrently). The transceiver 910 may also include a modem for: modulating a signal; providing the modulated signal for transmission (e.g., via one or more antennas 915, via a wired transmitter); receiving the modulated signal (e.g., from one or more antennas 915, from a wired receiver); and demodulating the signal. In some implementations, the transceiver 910 may include one or more interfaces, such as one or more interfaces coupled to the one or more antennas 915 configured to support various receiving or obtaining operations, or one or more interfaces coupled to the one or more antennas 915 configured to support various sending or outputting operations, or a combination thereof. In some implementations, the transceiver 910 may include or be configured to be coupled to one or more processors or memory components, which are operable to perform or support operations based on received or obtained information or signals, or to generate information or other signals for transmission or other output, or any combination thereof. In some implementations, the transceiver 910, or the transceiver 910 and one or more antennas 915, or the transceiver 910 and one or more antennas 915 and one or more processors or memory components (e.g., processor 935 or memory 925 or both) may be included in a chip or chip assembly installed in the device 905. In some examples, the transceiver may be operable to support communications via one or more communication links (eg, communication link 125 , backhaul communication link 120 , midhaul communication link 162 , fronthaul communication link 168 ).

[0166] The memory 925 may include RAM and ROM. The memory 925 may store computer-readable, computer-executable code 930 including instructions that, when executed by the processor 935, cause the device 905 to perform various functions described herein. The code 930 may be stored in a non-transitory computer-readable medium such as a system memory or another type of memory. In some cases, the code 930 may not be directly executable by the processor 935, but may (for example, when compiled and executed) cause the computer to perform the functions described herein. In some cases, the memory 925 may also include a BIOS, which may control basic hardware or software operations, such as interaction with peripheral components or devices, among other things.

[0167] The processor 935 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, an ASIC, a CPU, an FPGA, a microcontroller, a programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination thereof). In some cases, the processor 935 may be configured to operate a memory array using a memory controller. In some other cases, the memory controller may be integrated into the processor 935. The processor 935 may be configured to execute computer-readable instructions stored in a memory (e.g., a memory 925) to enable the device 905 to perform various functions (e.g., functions or tasks supporting techniques for reporting correlation metrics for machine learning reproducibility). For example, the device 905 or a component of the device 905 may include a processor 935 and a memory 925 coupled to the processor 935, and the processor 935 and the memory 925 are configured to perform various functions described herein. The processor 935 may be an example of a cloud computing platform (e.g., one or more physical nodes and supporting software such as an operating system, a virtual machine, or a container instance), which may host functions for performing functions of the device 905 (e.g., by executing code 930). The processor 935 may be any one or more suitable processors capable of executing scripts or instructions of one or more software programs stored in the device 905 (such as in the memory 925). In some specific implementations, the processor 935 may be a component of a processing system. A processing system may generally refer to a system or a series of machines or components that receive inputs and process these inputs to produce a set of outputs (which may be passed to, for example, other systems or components of the device 905). For example, the processing system of the device 905 may refer to a system including various other components or subcomponents of the device 905 (such as the processor 935, or the transceiver 910, or the communication manager 920, or other components or combinations of components of the device 905). The processing system of the device 905 may interface with other components of the device 905, and may process information (such as input or signals) received from other components or output information to other components. For example, a chip or modem of the device 905 may include a processing system and one or more interfaces for outputting information or for obtaining information or both. The one or more interfaces may be implemented as or otherwise include a first interface configured to output information and a second interface configured to obtain information or the same interface configured to output information and obtain information, as well as other specific implementations. In some specific implementations, the one or more interfaces may refer to an interface between a processing system of a chip or modem and a transmitter, so that the device 905 can send information output from the chip or modem. Additionally or alternatively, in some specific implementations, the one or more interfaces may refer to an interface between a processing system of a chip or modem and a receiver, so that the device 905 can obtain information or signal input, and the information can be passed to the processing system.One of ordinary skill in the art will readily recognize that the first interface may also obtain information or signal input, and the second interface may also output information or signal output.

[0168] In some examples, bus 940 may support communications of protocol layers (e.g., within a protocol layer) of a protocol stack. In some examples, bus 940 may support communications associated with logical channels of a protocol stack (e.g., between protocol layers of a protocol stack), which may include communications performed within components of device 905, or communications performed between different components of device 905 that may be co-located or may be located in different locations (e.g., where device 905 may refer to a system in which one or more of communication manager 920, transceiver 910, memory 925, code 930, and processor 935 may be located in one component of different components or may be divided between different components).

[0169] In some examples, the communication manager 920 may manage aspects of communications with the core network 130 (e.g., via one or more wired or wireless backhaul links). For example, the communication manager 920 may manage the transmission of data communications for client devices such as one or more UEs 115. In some examples, the communication manager 920 may manage communications with other network entities 105 and may include a controller or scheduler for controlling communications with the UEs 115 in coordination with the other network entities 105. In some examples, the communication manager 920 may support an X2 interface within an LTE / LTE-A wireless communication network technology to provide communications between network entities 105.

[0170] According to examples as disclosed herein, the communication manager 920 may support wireless communications at a machine learning server. For example, the communication manager 920 may be configured to or otherwise support a component for generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data. The communication manager 920 may be configured to or otherwise support a component for receiving one or more messages from one or more devices, the one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The communication manager 920 may be configured to or otherwise support a component for sending a message to the one or more devices, the message indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters.

[0171] Additionally or alternatively, according to examples as disclosed herein, the communication manager 920 may support wireless communication at the first device. For example, the communication manager 920 may be configured to or otherwise support components for generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The communication manager 920 may be configured to or otherwise support components for sending a first message to a machine learning server, the first message indicating the second set of parameters representing the test data. The communication manager 920 may be configured to or otherwise support components for receiving a second message indicating a reproducibility measure based on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimension relative to the dimension of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0172] By including or configuring a communication manager 920 according to examples as described herein, the device 905 may support techniques for reporting relevance metrics for machine learning reproducibility that may provide improved communication reliability, an improved user experience associated with reduced processing, and more efficient utilization of communication resources.

[0173] In some examples, the communication manager 920 may be configured to perform various operations (e.g., receive, obtain, monitor, output, transmit) using or otherwise cooperating with the transceiver 910, the one or more antennas 915 (e.g., where applicable), or any combination thereof. Although the communication manager 920 is illustrated as a separate component, in some examples, one or more functions described with reference to the communication manager 920 may be supported or performed by the transceiver 910, the processor 935, the memory 925, the code 930, or any combination thereof. For example, the code 930 may include instructions executable by the processor 935 to cause the device 905 to perform various aspects of the techniques for reporting correlation metrics for machine learning reproducibility as described herein, or the processor 935 and the memory 925 may be otherwise configured to perform or support such operations.

[0174] Fig.10 A flowchart illustrating a method 1000 of supporting techniques for reporting relevance metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The operations of the method 1000 may be implemented by a machine learning server, UE, or network entity, or components thereof, as described herein. For example, the operations of the method 1000 may be implemented by a machine learning server, UE, or network entity, or components thereof, as described herein. Figures 1 to 9The UE 115 or network entity described may be executed. In some examples, the UE or network entity may execute an instruction set to control the functional elements of the UE or network entity to perform the described functions. Additionally or alternatively, the UE or network entity may use dedicated hardware to perform various aspects of the described functions.

[0175] At 1005, the method may include generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimensionality relative to the dimensionality of the training data. The operations of 1005 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1005 may be performed as described in reference Figure 7 The training parameter component 725 is used to perform the above operation.

[0176] At 1010, the method may include receiving one or more messages from one or more devices indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment, or both, wherein the second set of parameters has a reduced dimension relative to a dimension of the test data. The operations of 1010 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1010 may be performed as described in reference to Figure 7 The test parameter manager 730 is used to execute.

[0177] At 1015, the method may include sending a message to the one or more devices indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters. The operations of 1015 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1015 may be performed as described in reference to Figure 7 The reproducibility measurement component 735 is used to perform the above.

[0178] Fig.11 A flowchart illustrating a method 1100 of supporting techniques for reporting relevance metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The operations of the method 1100 may be implemented by a machine learning server, UE, or network entity, or components thereof, as described herein. For example, the operations of the method 1100 may be implemented by a machine learning server, UE, or network entity, or components thereof, as described herein. Figures 1 to 9 The UE 115 or network entity described may be executed. In some examples, the UE or network entity may execute an instruction set to control the functional elements of the UE or network entity to perform the described functions. Additionally or alternatively, the UE or network entity may use dedicated hardware to perform various aspects of the described functions.

[0179] At 1105, the method may include generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimensionality relative to the dimensionality of the training data. The operations of 1105 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1105 may be performed as described in reference Figure 7 The training parameter component 725 is used to perform the above operation.

[0180] At 1110, the method may include receiving one or more messages from one or more devices indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to a dimension of the test data. The operations of 1110 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1110 may be performed as described in reference to Figure 7 The test parameter manager 730 is used to execute.

[0181] At 1115, the method may include generating a reproducibility metric based on a correlation between the first set of parameters and the second set of parameters. The operations of 1115 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1115 may be performed as described in reference Figure 7 The reproducibility measurement component 735 is used to perform the above.

[0182] At 1120, the method may include sending a message to the one or more devices indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based on a correlation between the first set of parameters and the second set of parameters. The operations of 1120 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1120 may be performed as described in reference to Figure 7 The reproducibility measurement component 735 is used to perform the above.

[0183] Fig.12 A flowchart illustrating a method 1200 of supporting techniques for reporting relevance metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The operations of the method 1200 may be implemented by a machine learning server, UE, or network entity, or components thereof, as described herein. For example, the operations of the method 1200 may be implemented by a machine learning server, UE, or network entity, or components thereof, as described herein. Figures 1 to 9 The UE 115 or network entity described may be executed. In some examples, the UE or network entity may execute an instruction set to control the functional elements of the UE or network entity to perform the described functions. Additionally or alternatively, the UE or network entity may use dedicated hardware to perform various aspects of the described functions.

[0184] At 1205, the method may include generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data. The operations of 1205 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1205 may be performed as described in reference Figure 7 The test parameter component 740 is used to execute.

[0185] At 1210, the method may include sending a first message to the machine learning server indicating a second set of parameters representing the test data. The operations of 1210 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1210 may be performed as described in reference to Figure 7 The test parameter manager 730 is used to execute.

[0186] At 1215, the method may include receiving a second message indicating a reproducibility metric based on a correlation between a first set of parameters and a second set of parameters, wherein the first set of parameters represents training data and has a reduced dimensionality relative to a dimensionality of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments, or a combination thereof. The operations of 1215 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1215 may be performed as described in reference Figure 7 The reproducibility metric manager 745 is used to perform the above operations.

[0187] Fig.13 A flowchart illustrating a method 1300 of supporting techniques for reporting relevance metrics for machine learning reproducibility according to one or more aspects of the present disclosure is illustrated. The operations of the method 1300 may be implemented by a machine learning server 210, a UE, or a network entity or components thereof as described herein. For example, the operations of the method 1300 may be implemented by a machine learning server 210, a UE, or a network entity as described herein. Figures 1 to 9 The UE 115 or network entity described may be performed. In some examples, the UE or network entity may execute an instruction set to control the functional elements of the UE or network entity to perform the described functions. Additionally or alternatively, the UE or network entity may use dedicated hardware to perform various aspects of the described functions.

[0188] At 1305, the method may include receiving a message indicating one or more parameters from a device within the communication environment or the channel environment or both. The operations of 1305 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1305 may be performed as described in reference to Figure 7 The collaboration component 755 is used to execute.

[0189] At 1310, the method may include generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data, and wherein generating the second set of parameters is based on receiving a message indicating the one or more parameters. The operations of 1310 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1310 may be performed as described in reference to Figure 7 The test parameter component 740 is used to execute.

[0190] At 1315, the method may include sending a first message to the machine learning server indicating a second set of parameters representing the test data. The operations of 1315 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1315 may be performed as described in reference to Figure 7 The test parameter manager 730 is used to execute.

[0191] At 1320, the method may include receiving a second message indicating a reproducibility metric based on a correlation between a first set of parameters and a second set of parameters, wherein the first set of parameters represents training data and has a reduced dimensionality relative to a dimensionality of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments, or a combination thereof. The operations of 1320 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1320 may be performed as described in reference to Figure 7 The reproducibility metric manager 745 is used to perform the above operations.

[0192] The following provides an overview of various aspects of the disclosure:

[0193] Aspect 1: A method for wireless communication at a machine learning server, the method comprising: generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimension relative to the dimension of the training data; receiving one or more messages from one or more devices indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimension relative to the dimension of the test data; and sending a message to the one or more devices indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based at least in part on a correlation between the first set of parameters and the second set of parameters.

[0194] Aspect 2: A method according to Aspect 1, wherein the reproducibility metric indicates to the one or more devices to perform one or more communication procedures based on one or more predictions from the machine learning model generated by the machine learning server using the training data, and the reproducibility metric indicates to perform the one or more communication procedures based at least in part on the reproducibility metric satisfying a threshold.

[0195] Aspect 3: A method according to any one of Aspects 1 to 2, wherein the reproducibility metric indicates to the one or more devices that one or more communication procedures are to be performed based on one or more measurements performed by the one or more devices, and the reproducibility metric indicates that the one or more communication procedures are to be performed based at least in part on the reproducibility metric failing to meet a threshold.

[0196] Aspect 4: A method according to any one of Aspects 1 to 3, wherein the reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by the machine learning server using the training data applied to the communication environment.

[0197] Aspect 5: The method according to Aspect 4, wherein the prediction from the machine learning model generated by the machine learning server includes beam weight predictions for non-blocking scenarios or blocking scenarios.

[0198] Aspect 6: The method according to any one of aspects 1 to 5, further comprising: generating the reproducibility metric based at least in part on the correlation between the first set of parameters and the second set of parameters.

[0199] Aspect 7: A method according to Aspect 6, wherein generating the reproducibility metric includes: generating the reproducibility metric based at least in part on the correlation between one or more principal components associated with the training data and one or more principal components associated with the test data, wherein the first set of parameters includes the one or more principal components associated with the training data, and wherein the second set of parameters includes the one or more principal components associated with the test data.

[0200] Aspect 8: According to the method of any one of Aspects 1 to 7, the method also includes: generating additional training data for the one or more communication environments or the one or more channel environments or a combination thereof based at least in part on the reproducibility metric failing to meet a threshold.

[0201] Aspect 9: A method according to any one of Aspects 1 to 8, wherein the training data includes an indication of at least one of the following: the number of clusters in the one or more communication environments or the one or more channel environments or a combination thereof, the number of reflectors in the one or more communication environments or the one or more channel environments or a combination thereof, the number of diffraction points in the one or more communication environments or the one or more channel environments or a combination thereof, positioning information associated with one or more devices in the one or more communication environments or the one or more channel environments or a combination thereof, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

[0202] Aspect 10: A method according to any one of Aspects 1 to 9, wherein the test data includes an indication of at least one of the following: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the one or more devices, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

[0203] Aspect 11: A method according to any one of Aspects 1 to 10, wherein the machine learning server is part of a UE or part of a network entity, or comprises an independent node associated with multiple devices in a network.

[0204] Aspect 12: A method for performing wireless communications at a first device, the method comprising: generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimensionality relative to the dimensionality of the test data; sending a first message indicating the second set of parameters representing the test data to a machine learning server; and receiving a second message indicating a reproducibility metric based at least in part on a correlation between the first set of parameters and the second set of parameters, wherein the first set of parameters represents training data and has a reduced dimensionality relative to the dimensionality of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

[0205] Aspect 13: A method according to Aspect 12, wherein the reproducibility metric indicates that one or more communication procedures are performed based on one or more predictions from a machine learning model generated by the machine learning server using the training data, and the reproducibility metric indicates that the first device performs the one or more communication procedures at least in part based on the reproducibility metric satisfying a threshold.

[0206] Aspect 14: A method according to any one of Aspects 12 to 13, wherein the reproducibility metric indicates that one or more communication procedures are performed based on one or more measurements performed by the first device, and the reproducibility metric indicates that the first device performs the one or more communication procedures at least in part based on the reproducibility metric failing to meet a threshold.

[0207] Aspect 15: A method according to any one of Aspects 12 to 14, wherein the reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by the machine learning server using the training data applied to the communication environment or the channel environment or both.

[0208] Aspect 16: A method according to Aspect 15, wherein the prediction from the machine learning model associated with the training data generated by the machine learning server includes beam weight predictions for non-blocking scenarios or blocking scenarios.

[0209] Aspect 17: According to the method described in any one of Aspects 12 to 16, the method also includes: receiving a message indicating one or more parameters from a second device within the communication environment or the channel environment or both, wherein generating the second set of parameters is at least partially based on receiving the message indicating the one or more parameters.

[0210] Aspect 18: The method according to any one of aspects 12 to 17, further comprising: sending a message indicating the reproducibility metric to a second device within the communication environment or the channel environment or both based at least in part on receiving the reproducibility metric.

[0211] Aspect 19: A method according to any one of Aspects 12 to 18, wherein the test data includes an indication of at least one of the following: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the first device, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the first device, or the beam measurement capability of the first device.

[0212] Aspect 20: A method according to any one of Aspects 12 to 19, wherein the training data includes an indication of at least one of the following: the number of clusters in the one or more communication environments or the one or more channel environments or a combination thereof, the number of reflectors in the one or more communication environments or the one or more channel environments or a combination thereof, the number of diffraction points in the one or more communication environments or the one or more channel environments or a combination thereof, positioning information associated with one or more devices in the one or more communication environments or the one or more channel environments or a combination thereof, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

[0213] Aspect 21: A device for wireless communication at a machine learning server, the device comprising: a processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the device to perform a method according to any one of Aspects 1 to 11.

[0214] Aspect 22: An apparatus for wireless communication at a machine learning server, the apparatus comprising: at least one component for performing a method according to any one of Aspects 1 to 11.

[0215] Aspect 23: A non-transitory computer-readable medium storing code for wireless communication at a machine learning server, the code comprising instructions executable by a processor to perform a method according to any one of aspects 1 to 11.

[0216] Aspect 24: An apparatus for wireless communication at a first device, the apparatus comprising: a processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform a method according to any one of Aspects 12 to 20.

[0217] Aspect 25: An apparatus for wireless communication at a first device, the apparatus comprising: at least one component for performing the method according to any one of aspects 12 to 20.

[0218] Aspect 26: A non-transitory computer-readable medium storing code for wireless communication at a first device, the code comprising instructions executable by a processor to perform the method according to any one of aspects 12 to 20.

[0219] It should be noted that the methods described herein describe possible implementations, and that the various operations and steps may be rearranged or otherwise modified and that other implementations are possible. Furthermore, aspects from two or more methods may be combined.

[0220] Although aspects of an LTE, LTE-A, LTE-A Pro, or NR system may be described for example purposes, and LTE, LTE-A, LTE-A Pro, or NR terminology may be used in much of the description, the techniques described herein may also be applicable to networks other than LTE, LTE-A, LTE-A Pro, or NR networks. For example, the described techniques may be applicable to various other wireless communication systems, such as Ultra Mobile Broadband (UMB), Institute of Electrical and Electronics Engineers (IEEE) 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, Flash-OFDM, and other systems and radio technologies not explicitly mentioned herein.

[0221] The information and signals described herein may be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the specification may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or optical particles, or any combination thereof.

[0222] The various illustrative blocks and components described in conjunction with the disclosure herein may be implemented or executed using a general purpose processor, DSP, ASIC, CPU, FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in an alternative, the processor may be any processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration).

[0223] The functions described herein may be implemented using hardware, software executed by a processor, firmware, or any combination thereof. If implemented using software executed by a processor, the functions may be stored as one or more instructions or codes of a computer-readable medium, or sent using one or more instructions or codes of a computer-readable medium. Other examples and specific implementations are within the scope of this disclosure and the appended claims. For example, due to the nature of software, the functions described herein may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these items. Features that implement the functions may also be physically located at different locations, including being distributed so that the various parts of the functions are implemented at different physical locations.

[0224] Computer-readable media include both non-transient computer storage media and communication media, which include any media that facilitates the transfer of computer programs from one location to another. Non-transient storage media can be any available media that can be accessed by a general or special computer. By way of example and not limitation, non-transient computer-readable media may include RAM, ROM, electrically erasable programmable ROM (EEPROM), flash memory, compact disc (CD) ROM or other optical disc storage devices, magnetic disk storage devices or other magnetic storage devices, or may be used to carry or store desired program code components in the form of instructions or data structures and any other non-transient media that can be accessed by a general or special computer or a general or special processor. Moreover, any connection is appropriately referred to as a computer-readable medium. For example, if the software is sent from a website, server or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL) or wireless technologies such as infrared, radio and microwaves, coaxial cable, fiber optic cable, twisted pair, DSL or wireless technologies such as infrared, radio and microwaves are included in the definition of computer-readable media. As used herein, disks and optical disks include CDs, laser disks, optical disks, digital versatile disks (DVDs), floppy disks, and Blu-ray disks. Disks can reproduce data magnetically, and optical disks can reproduce data optically using lasers. Combinations of the above are also included within the scope of computer-readable media.

[0225] As used herein (including in the claims), "or" used in a list of items (e.g., a list of items followed by a phrase such as "at least one of" or "one or more of") indicates an inclusive list, so that, for example, a list of at least one of A, B, or C means A or B or C or AB or AC or BC or ABC (i.e., A and B and C). In addition, as used herein, the phrase "based on" should not be interpreted as a reference to a closed set of conditions. For example, an example step described as "based on condition A" can be based on both condition A and condition B without departing from the scope of the present disclosure. In other words, as used herein, the phrase "based on" should be interpreted in the same manner as the phrase "based at least in part on".

[0226] The term "determining" encompasses a variety of actions, and thus, "determining" may include calculating, computing, processing, deriving, investigating, looking up (such as via looking up in a table, database or other data structure), ascertaining, and the like. Furthermore, "determining" may include receiving (e.g., receiving information), accessing (e.g., accessing data stored in a memory), etc. Additionally, "determining" may include parsing, obtaining, selecting, choosing, establishing, and other such similar actions.

[0227] In the drawings, similar components or features may have the same reference label. In addition, various components of the same type may be distinguished by following the reference label with a dash and a second label to distinguish between the similar components. If only the first reference label is used in the specification, the description may apply to any of the similar components having the same first reference label, regardless of the second reference label or other subsequent reference labels.

[0228] The descriptions set forth herein in conjunction with the accompanying drawings describe example configurations and do not represent all examples that may be implemented or within the scope of the claims. The term "example" as used herein means "used as an example, instance, or illustration," rather than "preferred" or "advantageous over other examples." The specific implementation includes specific details to provide an understanding of the described techniques. However, these techniques may be practiced without these specific details. In some cases, known structures and devices are shown in block diagram form to avoid obscuring the concepts of the described examples.

[0229] The description herein is provided to enable one of ordinary skill in the art to implement or use the present disclosure. Various modifications to the present disclosure will be apparent to one of ordinary skill in the art, and the general principles defined herein may be applied to other variations without departing from the scope of the present disclosure. Thus, the present disclosure is not limited to the examples and designs described herein, but should be granted the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An apparatus for wireless communication at a machine learning server, the apparatus comprising: processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to: generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimensionality relative to a dimensionality of the training data; receiving, from one or more devices, one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimensionality relative to a dimensionality of the test data; as well as A message is sent to the one or more devices indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based at least in part on a correlation between the first set of parameters and the second set of parameters.

2. An apparatus according to claim 1, wherein the reproducibility metric indicates to the one or more devices to perform one or more communication procedures based on one or more predictions from the machine learning model generated by the machine learning server using the training data, and the reproducibility metric indicates to perform the one or more communication procedures based at least in part on the reproducibility metric satisfying a threshold.

3. An apparatus according to claim 1, wherein the reproducibility metric indicates to the one or more devices to perform one or more communication procedures based on one or more measurements performed by the one or more devices, and the reproducibility metric indicates to perform the one or more communication procedures based at least in part on the reproducibility metric failing to meet a threshold.

4. An apparatus according to claim 1, wherein the reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by the machine learning server using the training data applied to the communication environment.

5. The apparatus of claim 4, wherein the predictions from the machine learning model generated by the machine learning server include beam weight predictions for a non-blocking scenario or a blocking scenario.

6. The apparatus of claim 1, wherein the instructions are further executable by the processor to cause the apparatus to: The reproducibility metric is generated based at least in part on the correlation between the first set of parameters and the second set of parameters.

7. The apparatus of claim 6, wherein the instructions for generating the reproducibility metric are executable by the processor to cause the apparatus to: The reproducibility metric is generated based at least in part on a correlation between one or more principal components associated with the training data and one or more principal components associated with the test data, wherein the first set of parameters includes the one or more principal components associated with the training data, and wherein the second set of parameters includes the one or more principal components associated with the test data.

8. The apparatus of claim 1, wherein the instructions are further executable by the processor to cause the apparatus to: Additional training data is generated for the one or more communication environments or the one or more channel environments, or a combination thereof based at least in part on the reproducibility metric failing to meet a threshold.

9. An apparatus according to claim 1, wherein the training data includes an indication of at least one of the following: the number of clusters in the one or more communication environments or the one or more channel environments or a combination thereof, the number of reflectors in the one or more communication environments or the one or more channel environments or a combination thereof, the number of diffraction points in the one or more communication environments or the one or more channel environments or a combination thereof, positioning information associated with one or more devices in the one or more communication environments or the one or more channel environments or a combination thereof, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

10. An apparatus according to claim 1, wherein the test data includes an indication of at least one of the following: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the one or more devices, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the one or more devices, or the beam measurement capability of the one or more devices.

11. The apparatus of claim 1, wherein the machine learning server is part of a user equipment (UE) or part of a network entity, or comprises an independent node associated with a plurality of devices in a network.

12. An apparatus for wireless communication at a first device, the apparatus comprising: processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to: generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimensionality relative to a dimensionality of the test data; sending a first message indicating the second set of parameters representing the test data to a machine learning server; as well as Receive a second message indicating a reproducibility metric based at least in part on a correlation between a first set of parameters and a second set of parameters, wherein the first set of parameters represents training data and has a reduced dimensionality relative to a dimensionality of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

13. An apparatus according to claim 12, wherein the reproducibility metric indicates that one or more communication procedures are performed based on one or more predictions from a machine learning model generated by the machine learning server using the training data, and the reproducibility metric indicates that the first device performs the one or more communication procedures at least in part based on the reproducibility metric satisfying a threshold.

14. An apparatus according to claim 12, wherein the reproducibility metric indicates that one or more communication procedures are performed based on one or more measurements performed by the first device, and the reproducibility metric indicates that the first device performs the one or more communication procedures at least in part based on the reproducibility metric failing to meet a threshold.

15. An apparatus according to claim 12, wherein the reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by the machine learning server using the training data applied to the communication environment or the channel environment or both.

16. The apparatus of claim 15, wherein the predictions from the machine learning model associated with the training data generated by the machine learning server include beam weight predictions for a non-blocking scenario or a blocking scenario.

17. The apparatus of claim 12, wherein the instructions are further executable by the processor to cause the apparatus to: A message indicating one or more parameters is received from a second device within the communication environment or the channel environment or both, wherein generating the second set of parameters is based at least in part on receiving the message indicating the one or more parameters.

18. The apparatus of claim 12, wherein the instructions are further executable by the processor to cause the apparatus to: A message indicative of the reproducibility metric is sent to a second device within the communication environment or the channel environment or both based at least in part on receiving the reproducibility metric.

19. An apparatus according to claim 12, wherein the test data includes an indication of at least one of the following: the number of clusters in the communication environment or the channel environment or both, the number of reflectors in the communication environment or the channel environment or both, the number of diffraction points in the communication environment or the channel environment or both, positioning information associated with the first device, antenna module capabilities, the number of antenna modules, the arrangement of antenna modules, the beamforming capability of the first device, or the beam measurement capability of the first device.

20. A method for wireless communication at a machine learning server, the method comprising: generating a first set of parameters representing training data, the training data corresponding to one or more communication environments or one or more channel environments or a combination thereof, wherein the first set of parameters has a reduced dimensionality relative to a dimensionality of the training data; receiving, from one or more devices, one or more messages indicating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimensionality relative to a dimensionality of the test data; as well as A message is sent to the one or more devices indicating a reproducibility metric of the communication environment or the channel environment or both, the reproducibility metric being based at least in part on a correlation between the first set of parameters and the second set of parameters.

21. A method according to claim 20, wherein the reproducibility metric indicates to the one or more devices to perform one or more communication procedures based on one or more predictions from the machine learning model generated by the machine learning server using the training data, and the reproducibility metric indicates to perform the one or more communication procedures based at least in part on the reproducibility metric satisfying a threshold.

22. A method according to claim 20, wherein the reproducibility metric indicates to the one or more devices to perform one or more communication procedures based on one or more measurements performed by the one or more devices, and the reproducibility metric indicates to perform the one or more communication procedures based at least in part on the reproducibility metric failing to meet a threshold.

23. A method according to claim 20, wherein the reproducibility metric indicates a confidence value associated with a prediction from a machine learning model generated by the machine learning server using the training data applied to the communication environment.

24. The method of claim 23, wherein the predictions from the machine learning model associated with the training data generated by the machine learning server include beam weight predictions for a non-blocking scenario or a blocking scenario.

25. The method according to claim 20, further comprising: Additional training data is generated for the one or more communication environments or the one or more channel environments, or a combination thereof based at least in part on the reproducibility metric failing to meet a threshold.

26. A method for wireless communication at a first device, the method comprising: generating a second set of parameters representing test data, the test data corresponding to a communication environment or a channel environment or both, wherein the second set of parameters has a reduced dimensionality relative to a dimensionality of the test data; sending a first message indicating the second set of parameters representing the test data to a machine learning server; as well as Receive a second message indicating a reproducibility metric based at least in part on a correlation between a first set of parameters and a second set of parameters, wherein the first set of parameters represents training data and has a reduced dimensionality relative to a dimensionality of the training data, wherein the training data corresponds to one or more communication environments or one or more channel environments or a combination thereof.

27. A method according to claim 26, wherein the reproducibility metric indicates that one or more communication procedures are performed based on one or more predictions from the machine learning model generated by the machine learning server using the training data, and the reproducibility metric indicates that the first device performs the one or more communication procedures at least in part based on the reproducibility metric satisfying a threshold.

28. A method according to claim 26, wherein the reproducibility metric indicates that one or more communication procedures are performed based on one or more measurements performed by the first device, and the reproducibility metric indicates that the first device performs the one or more communication procedures at least in part based on the reproducibility metric failing to meet a threshold.

29. The method according to claim 26, further comprising: A message indicating one or more parameters is received from a second device within the communication environment or the channel environment or both, wherein generating the second set of parameters is based at least in part on receiving the message indicating the one or more parameters.

30. The method of claim 26, further comprising: A message indicative of the reproducibility metric is sent to a second device within the communication environment or the channel environment or both based at least in part on receiving the reproducibility metric.

Images