Communication method and related apparatus

Abstract

The present application provides a communication method and a communication apparatus, which are applied to the technical field of communications. In the technical solution provided by the present application, a receiving node acquires an output result of a first artificial intelligence (AI) model, and interpretability analysis is additionally performed on the output result. The receiving node acquires an interpretable result of the output result, and determines the reasonableness of an inferred model output result. Thus, the addition of model interpretability analysis and interaction into an AI application in a communication network by the present application increases the reasonableness of an AI inference result, thereby ensuring the reliability of the application of an AI algorithm to the communication network.

Description

Communication method and related device

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on November 27, 2023, with application number 202311604208.X and application name “Communication Methods and Related Devices”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to communication methods and related devices. Background Art

[0003] With the development of artificial intelligence (AI) and the continuous improvement of the performance requirements of communication systems, AI technology is being applied in more and more scenarios in communication systems.

[0004] Currently, the technical solution for applying AI technology in communication systems is as follows: the receiving node initiates an AI task request or provides the data required for AI reasoning to the model management node; the model management node completes the AI ​​reasoning locally or jointly with the computing network element; the model management node feeds the reasoning results back to the receiving node; and the receiving node returns result confirmation or feedback information to the model management node. The first and fourth steps can each be optional steps. The receiving node is the requestor of AI reasoning, such as a terminal; the model management node is the provider and execution node of AI services in the communication network.

[0005] However, when applying AI technology in a communication system as described above, there are the following problems: the inference results of the AI ​​model fed back by the model management node to the receiving node lack reliability judgment, resulting in the inability to guarantee the reliability of the communication performance of the communication system. Summary of the Invention

[0006] This application provides a communication method and related devices that can achieve reliability feedback of the inference results of the AI ​​model in the communication system, thereby improving communication performance.

[0007] In a first aspect, the present application provides a communication method for a communication node, or a device, module, circuit, or chip used in a communication node, or a device capable of being used in conjunction with a communication node. The method comprises: obtaining an inference result of a first artificial intelligence (AI) model; and obtaining an interpretable result of the inference result.

[0008] In this method, an AI model interpretability mechanism is added to the communication node to obtain interpretable results of the AI ​​model's reasoning results. Because interpretable results can be used to explain the AI ​​model, they can be used to judge the rationality of the AI ​​model's reasoning results, and thus can be used to evaluate the reliability of the reasoning results after use. Therefore, obtaining interpretable results can provide technical support for judging the credibility and rationality of the AI ​​model's reasoning results, thereby helping to ensure the reliability of AI algorithms in communication systems and thus improving communication performance.

[0009] It can be understood that obtaining the interpretable result in the method may include: receiving the interpretable result, or the communication node determining the interpretable result itself.

[0010] In some possible implementations, the method may further include: sending first information indicating first interpretable related information, the first interpretable related information being AI model interpretable related information, the AI ​​model interpretable related information including at least one of the following information: an AI model interpretability method, an output type, an interpretation content of the AI ​​model interpretability method, or an interpretation accuracy of the AI ​​model interpretability method, the output type being the type of the output result of the AI ​​model. The first information is used by a communication node that receives the first information to determine an interpretable result of the AI ​​model's inference result.

[0011] In this implementation, optionally, obtaining the inference result of the AI ​​model may include receiving the inference result of the AI ​​model; and obtaining the explainable result may include receiving the explainable result. Sending the first information may occur before receiving the explainable result, and further, may occur before receiving the inference result.

[0012] In this implementation, the current communication node provides a model interpretability indicator to the communication node that provides the reasoning results of the AI ​​model, so that the communication node that receives the first information can obtain an interpretable result that better meets the needs of the current communication node, thereby improving the rationality of the model interpretability analysis.

[0013] In some possible implementations, the first information is one of a plurality of interpretable related information, wherein the first information includes an index of the first interpretable related information in the plurality of interpretable related information.

[0014] In other words, the current communication node indicates the interpretable related information required to determine the interpretable result to the communication node that determines the interpretable result through the index.

[0015] In some possible implementations, the method further includes: receiving second information, where the second information indicates interpretable related information associated with each output type of the at least one output type.

[0016] In other words, the current communication node learns the interpretable related information supported by the communication node that determines the interpretable result through the second information. In this case, when the current communication node sends the first information, the interpretable related information indicated by the first information may be determined by the second information. For example, the interpretable related information indicated by the first information may be the interpretable related information supported by the communication node that determines the interpretable result.

[0017] In some possible implementations, the method further includes: receiving third information, the third information indicating information required to obtain an interpretable result of the inference result of the first AI model; wherein, obtaining the interpretable result of the inference result includes: determining the interpretable result based on the third information.

[0018] Alternatively, the third information is used to obtain the input data required for interpretability analysis, and the interpretability analysis is performed using the input data to obtain an interpretable result. For example, the third information may be received when the current communication node determines the interpretable result itself.

[0019] In some possible implementations, the method further includes: sending fourth information, where the fourth information is used to request information required to obtain an interpretable result of the inference result of the first AI model.

[0020] Alternatively, the current communication node may request the input data required for interpretability analysis through the fourth information. For example, the fourth information may be sent when the current communication node determines an interpretable result on its own.

[0021] In some possible implementations, the method further includes: sending an interpretable result.

[0022] For example, when the interpretable result is used by other communication nodes to analyze the rationality of the reasoning result, the current communication node can send the interpretable result determined by itself to the other communication nodes.

[0023] In some possible implementations, obtaining an interpretable result of the inference result includes receiving the interpretable result. For example, if the interpretable result is obtained by analysis by other communication nodes, the current communication node can receive the interpretable result from the other communication nodes.

[0024] In some possible implementations, the method further includes: sending fifth information indicating the rationality of the inference result of the first AI model.

[0025] For example, when the current communication node is a node that uses explainable results, the current communication node can determine or judge the rationality of the reasoning result based on the explainable results, and feedback the rationality of the reasoning result to other communication nodes that determine the reasoning result, so that other communication nodes can adjust the AI ​​model and perform other processing to improve the reliability of the reasoning result, thereby improving communication reliability.

[0026] In a second aspect, the present application provides a communication device, comprising modules or units for implementing the method in the first aspect and any possible implementation of the first aspect. It should be understood that each module or unit can implement the corresponding function by executing a computer program.

[0027] In a third aspect, the present application provides a communication device, comprising a processor, wherein the processor is configured to execute the communication method described in the first aspect or any possible implementation of the first aspect. The communication device may be a chip or chip system used in a terminal device.

[0028] The apparatus may further include a memory for storing instructions and data. The memory is coupled to the processor, and when the processor executes the instructions stored in the memory, the method described in the first aspect or any possible implementation thereof may be implemented. The apparatus may further include a communication interface for communicating between the apparatus and other devices. Exemplarily, the communication interface may be a transceiver, circuit, bus, module, or other type of communication interface.

[0029] In a fourth aspect, the present application provides a computer-readable storage medium storing a program code for execution by a communication device, wherein the program code includes instructions for implementing the method in the first aspect and any possible implementation manner of the first aspect.

[0030] In a fifth aspect, the present application provides a computer program product comprising instructions, which, when executed on a communication device, enables the communication device to implement the method in the first aspect and any possible implementation manner of the first aspect.

[0031] In a sixth aspect, the present application provides a communication system, which includes a communication device for implementing the method in the first aspect and any possible implementation manner of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] FIG1 is a schematic diagram of a communication system applicable to the method of an embodiment of the present application;

[0033] FIG2 is a schematic diagram of another communication system applicable to the method of an embodiment of the present application;

[0034] FIG3 is a schematic flow chart of a communication method provided in one embodiment of the present application;

[0035] FIG4 is a schematic flow chart of a communication method provided in one embodiment of the present application;

[0036] FIG5 is a schematic structural diagram of a communication device according to an embodiment of the present application;

[0037] FIG6 is a schematic structural diagram of a communication device provided in one embodiment of the present application;

[0038] FIG7 is a schematic diagram of the structure of a communication device provided by an embodiment of the present application;

[0039] FIG8 is a schematic structural diagram of a communication device provided in yet another embodiment of the present application. DETAILED DESCRIPTION

[0040] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0041] To facilitate the clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the words "first" and "second" are used to distinguish between identical or similar items with substantially the same functions and effects. Those skilled in the art will understand that the words "first" and "second" do not limit the quantity or execution order, and the words "first" and "second" do not necessarily mean different.

[0042] It should be noted that in the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described in this application as "exemplary" or "for example" should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0043] In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b and (or) c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0044] To facilitate understanding of the communication method provided in the embodiments of the present application, the system architecture and application scenarios of the communication method provided in the embodiments of the present application are described below. It is understood that the system architecture and application scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application.

[0045] In recent years, due to the rapid development of machine learning, especially deep learning technology, AI's capabilities in prediction, fitting, classification, and online autonomous decision-making have continued to improve. While strengthening many traditional applications, it has also brought about more emerging scenarios. For mobile systems, machine learning technology has been considered for application in the physical layer to replace traditional modules such as channel prediction. In addition, in application scenarios involving inter-node interaction, AI has also brought more methods for predicting communication events or behaviors, such as load balancing and mobility optimization. In response to problems such as high power consumption introduced by 5G, various AI methods based on deep learning (DL) or reinforcement learning (RL) have been introduced to formulate reasonable energy-saving strategies.

[0046] It should be noted that the multi-node AI application interaction involved in the communication system of this application includes but is not limited to channel prediction, mobility optimization, load balancing, and AI prediction information interaction between base stations. This application does not focus on the AI ​​application scenarios in the communication system. As long as it is a communication system that can implement the functional set of AI applications in the communication system, it should be included in the scope of protection of this application.

[0047] The communication field combines the application of AI technology. The technical solutions provided by this application can be applied to various communication systems, such as: fifth-generation (5G) or new radio (NR) systems, long-term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, wireless local area networks (WLAN) systems, satellite communication systems, future communication systems, such as sixth-generation (6G) mobile communication systems, or integrated systems of multiple systems. The technical solutions provided by this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine type communication (MTC), and Internet of Things (IoT) communication systems or other communication systems, such as the sixth generation (6G).

[0048] A device in a communication system can send a signal to another device or receive a signal from another device. The signal may include information, signaling, or data, etc. The device can also be replaced by an entity, a network entity, a communication device, a communication module, a node, a communication node, etc. The present application takes a communication node as an example for description. For example, a communication system may include multiple communication nodes, such as at least one receiving node and a model management node. The model management node can send a signal to the receiving node, and / or the receiving node can send a signal to the model management node. The receiving node can be the requester of AI reasoning, responsible for receiving, analyzing and using the AI ​​reasoning results; the model management node can be the provision and execution node of AI services in the communication network, responsible for storing, training and managing the AI ​​model of the corresponding task, and responding to the request of the receiving node to perform reasoning on the AI ​​task.

[0049] In an embodiment of the present application, the receiving node may be a terminal device or a network device.

[0050] Among them, the terminal device can also be called user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device.

[0051] The terminal device may be a device that provides voice / data, such as a handheld device or vehicle-mounted device with a wireless connection function. At present, some examples of terminals are: mobile phones, tablet computers, laptop computers, PDAs, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, wearable devices, terminal devices in 5G networks or future evolved public land mobile communication networks (PLMNs). The terminal equipment in the network (PLMN), etc., is not limited to this in the embodiments of the present application.

[0052] As an example and not a limitation, in the embodiment of the present application, the terminal device may also be a wearable device. Wearable devices may also be called wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not only hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0053] In the embodiments of the present application, the device for realizing the function of the terminal device can be a terminal device, or a device capable of supporting the terminal device to realize the function, such as a chip system, which can be installed in the terminal device or used in combination with the terminal device. In the embodiments of the present application, the chip system can be composed of a chip, or it can include a chip and other discrete devices. In the embodiments of the present application, only the terminal device is used as an example for description, and the embodiments of the present application are not limited to the solutions of the embodiments of the present application.

[0054] The model management node in the embodiments of the present application may be a network device, wherein a network device is a device for communicating with a terminal device, and the network device may also be referred to as an access network device or a radio access network device, such as a base station. The network device in the embodiments of the present application may refer to a radio access network (RAN) node (or device) that connects the terminal device to a wireless network. A base station may broadly cover various names as follows, or replace the following names, such as: NodeB, evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), master station, secondary station, multi-standard radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), radio unit (RU), positioning node, etc. A base station may be a macro base station, a micro base station, a relay node, a donor node, or the like, or a combination thereof. The base station may also refer to a communication module, modem or chip that is set in the aforementioned equipment or device. The base station may also be a mobile switching center and a device that performs the base station function in D2D, V2X, and M2M communications, a network side device in a 6G network, a device that performs the base station function in future communication systems, etc. The base station can support networks with the same or different access technologies. Optionally, the RAN node may also be a server, a wearable device, a vehicle or an on-board device, etc. For example, the access network device in the vehicle to everything (V2X) technology may be a road side unit (RSU). The embodiments of the present application do not limit the specific technology and specific device form adopted by the network equipment.In some deployments, the network devices mentioned in the embodiments of the present application may include a CU, a DU, or both a CU and a DU, or a control plane CU node (central unit-control plane (CU-CP)), a user plane CU node (central unit-user plane (CU-UP)), and a DU node. For example, the network devices may include a gNB-CU-CP, a gNB-CU-UP, and a gNB-DU.

[0055] In some deployments, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes implementing portions of the base station's functionality. For example, a RAN node can be a CU, DU, CU-CP, CU-UP, or RU. The CU and DU can be separate or included in the same network element, such as the BBU. The RU can be included in a radio frequency device or radio unit, such as an RRU, AAU, or RRH.

[0056] The RAN node may support one or more types of fronthaul interfaces, with different fronthaul interfaces corresponding to DUs and RUs with different functions. If the fronthaul interface between the DU and the RU is a common public radio interface (CPRI), the DU is configured to implement one or more baseband functions, and the RU is configured to implement one or more radio frequency functions. If the fronthaul interface between the DU and the RU is another type of interface, relative to the CPRI, some of the downlink and / or uplink baseband functions, such as precoding, digital beamforming (BF), or one or more of inverse fast Fourier transform (IFFT) / cyclic prefix (CP) for downlink, are moved from the DU to the RU for implementation; and for uplink, one or more of digital beamforming (BF), or fast Fourier transform (FFT) / cyclic prefix (CP) removal, are moved from the DU to the RU for implementation. In one possible implementation, the interface may be an enhanced common public radio interface (eCPRI). In the eCPRI architecture, the division between the DU and RU is different, corresponding to different types (category, Cat) of eCPRI, such as eCPRI Cat A, B, C, D, E, and F.

[0057] Taking eCPRI Cat A as an example, for downlink transmission, based on layer mapping, the DU is configured to implement layer mapping and one or more functions preceding it (i.e., one or more of coding, rate matching, scrambling, modulation, and layer mapping). Other functions after layer mapping (e.g., RE mapping, digital beamforming (BF), or one or more of inverse fast Fourier transform (IFFT) / cyclic prefix (CP) addition) are moved to the RU for implementation. For uplink transmission, based on RE demapping, the DU is configured to implement demapping and one or more functions preceding it (i.e., one or more of decoding, rate matching, descrambling, demodulation, inverse discrete Fourier transform (IDFT), channel equalization, and RE demapping). Other functions after demapping (e.g., one or more of digital BF or fast Fourier transform (FFT) / CP removal) are moved to the RU for implementation. It is understandable that for the functional description of DU and RU corresponding to various types of eCPRI, reference can be made to the eCPRI protocol, which will not be described in detail here.

[0058] In one possible design, the processing unit for implementing baseband functions in the BBU is called a baseband high layer (BBH) unit, and the processing unit for implementing baseband functions in the RRU / AAU / RRH is called a baseband low layer (BBL) unit.

[0059] In different systems, CU (or CU-CP and CU-UP), DU or RU may also have different names, but those skilled in the art can understand their meanings. For example, in the ORAN system, CU may also be called O-CU (Open CU), DU may also be called O-DU, CU-CP may also be called O-CU-CP, CU-UP may also be called O-CU-UP, and RU may also be called O-RU. Any unit of CU (or CU-CP, CU-UP), DU and RU in this application can be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0060] In the embodiments of the present application, the device for implementing the functions of the network device can be a network device; it can also be a device that can support the network device to implement the functions, such as a chip system, a hardware circuit, a software module, or a hardware circuit and a software module. The device can be installed in the network device or used in conjunction with the network device. In the embodiments of the present application, only the device for implementing the functions of the model management node is used as a network device as an example for description, and does not constitute a limitation on the embodiments of the present application.

[0061] The network device and / or terminal device can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; it can also be deployed on the water surface; it can also be deployed on aircraft, balloons and satellites in the air. The embodiments of this application do not limit the scenarios in which the network device and the terminal device are located. In addition, the terminal device and the network device can be hardware devices, or they can be software functions running on dedicated hardware, software functions running on general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities including dedicated or general-purpose hardware devices and software functions. This application does not limit the specific forms of the terminal device and the network device.

[0062] Figure 1 is a schematic diagram of a communication system applicable to the method of an embodiment of the present application. As shown in Figure 1, the communication system may include a receiving node and a model management node.

[0063] The model management node can be simply referred to as the management node, which can be a base station with computing power or a computing network element in the network; the receiving node can be a terminal or a base station.

[0064] The management node is responsible for model training and reasoning on requested tasks, while the receiving node initiates and receives the inference results of AI tasks. After completing AI task inference and feeding back the inference results to the receiving node, the management node can perform model interpretability analysis and inform the receiving node of the interpretability analysis results. Alternatively, the receiving node can perform model interpretability analysis itself to obtain the interpretability analysis results, thereby assisting the receiving node in judging the inference results and subsequent operations.

[0065] Figure 2 is a schematic diagram of another communication system applicable to the method of an embodiment of the present application. As shown in Figure 2, compared with the communication system shown in Figure 1, the communication system may further include a third node.

[0066] The third node can be a dedicated model management network element for AI model evaluation. For example, the third node can be a base station with computing power or a computing network element in the network. One difference between the third node and the management node is that the third node is not the node that performs AI inference for the task.

[0067] In the communication system shown in FIG2 , a third-party node can perform model interpretability analysis and inform the receiving node of the interpretability analysis results to assist the receiving node in judging the inference results and performing subsequent operations.

[0068] In different scenarios where the communication system and AI technology are integrated, as shown in Figures 1 and 2, the AI ​​model can perform different functions. For example, when the receiving node is a UE and the model management node is a gNB, the UE will request AI tasks such as resource scheduling, prediction instructions, and perception from the gNB. When the receiving node is a gNB and the model management node is a gNB, in this cross-site scenario, the AI ​​model can be applied to mobility enhancement, handover prediction, and terminal behavior prediction. When the receiving node is a gNB and the model management node is a NodeC, it can be applied to most RAN AI scenarios, with the base station requesting AI inference results from the computing network element.

[0069] It should be noted that Figures 1 and 2 are simplified schematic diagrams for ease of understanding. In actual applications, the communication system may include multiple receiving nodes and multiple model management nodes. The embodiments of the present application do not limit the number of receiving nodes and model management nodes included in the communication system.

[0070] Most current mainstream AI models utilize an end-to-end training and inference model, which creates a "black box effect." There's no unified, universal explanation for how training data guides AI model updates, or how AI models understand and process input data. Furthermore, complex models and large models with general capabilities are becoming increasingly mainstream. However, there's a negative correlation between model complexity and interpretability, making more complex models more difficult to interpret, leading to increased uncertainty about the credibility of their output.

[0071] Therefore, explainable AI (XAI) has also become a hot topic of discussion in academia and industry. Explainable methods refer to methods that understand and trust the reasoning / training process and results of machine learning models, thereby helping to describe the accuracy, fairness, transparency, and rationality of the models.

[0072] Many existing XAI methods offer a wide range of options for model interpretation, such as triggering from dimensions like text, images, and feature importance, to generate AI explainable solutions that are understandable to humans. XAI can be categorized into innate and acquired interpretability based on the inherent interpretability of the model. The former refers to models whose mechanisms and design inherently provide interpretability, such as decision trees, feature decomposition, and support vector machines (SVMs). The latter requires additional interpretability analysis of the model and is generally applicable to deep learning models such as neural networks (NNs) and transformers. Among these acquired, model-agnostic XAI methods, popular approaches include local interpretable model-agnostic explanations (Lime), which trains interpretable equivalent linear models using local samples; Shapley additive explanations (SHAP), which are explanations based on the Shapley value of each feature; and the Anchors method, which finds the minimum interpretable subset of features.

[0073] Regarding AI methods of interest to mobile networks, multiple use cases currently introduced in NG-RAN AI involve interactions between different nodes. These interactions encompass AI model inputs, outputs, and inference strategies. The AI ​​models used by different nodes are typically internally implemented, meaning they don't interact directly across vendors. Therefore, XAI methods are necessary to interpret the models and their outputs.

[0074] An exemplary application scenario of the communication method of the present application is to use the XAI method to obtain interpretable results of the prediction and reasoning results of the AI ​​model when the prediction and reasoning results of the AI ​​model are interacted between communication nodes, and then judge the rationality of the prediction and reasoning results of the AI ​​model, thereby ensuring the reliability of communication.

[0075] The following will introduce the new communication method proposed in this application in conjunction with specific embodiments. By adapting the XAI method to AI applications in mobile networks, an interpretable analysis of the model in the interaction is provided, thereby helping the receiving node to judge the reliability of the inference node output.

[0076] In a communication method provided in one embodiment of the present application, a management node performs reasoning and model interpretability analysis of AI tasks, and sends the reasoning results and interpretable results to the receiving node, so that the receiving node can determine the credibility of the reasoning results and perform subsequent tasks. An exemplary flowchart of the communication method of this embodiment is shown in Figure 3. The method may include S305, S310, S320, S330, and S340.

[0077] S305: The model management node sends second information to the receiving node, where the second information is used to indicate interpretable related information associated with each output type in the at least one output type. Accordingly, the receiving node receives the second information.

[0078] As an example, the output type may include at least one of the following types: the type of AI inference result that the model management node subsequently provides to the receiving node, for example, the model management node subsequently provides a prediction result or performs a measurement for the receiving node; specific data that the model management node needs to collect from the receiving node; the model management node instructs the receiving node to change the currently applied AI model strategy; or, a black box AI model.

[0079] As an example, the explainable related information may include at least one of the following information: an explainability method, which is used to indicate the method for performing explainability analysis; the explanation content of the explainability method, which is used to indicate the explanation content corresponding to different explainability methods; and the accuracy of the explainability method, which is used to indicate the explanation accuracy provided by different explainability methods.

[0080] As an example, the interpretable related information may include an output type. In this case, it means that the interpretable related information is associated with the output type, or in other words, other information included in the interpretable related information is associated with the output type.

[0081] The interpretability-related information associated with the output type includes the interpretability method, which can be understood as: the interpretability method recommended or supported by the AI ​​reasoning results of this output type, or the interpretability method supported or recommended by the model management node.

[0082] In some possible implementations, the second information can be understood as the model management node sending an inference type indication to the receiving node and requesting confirmation of the explainability method. Alternatively, the management node sending the second information can be understood as the management node initiating an inference indication and XAI confirmation.

[0083] As an example, explainability methods may include but are not limited to the following methods: Lime, Shap, Anchors, etc.

[0084] For example, different interpretability methods return different explanations. For example, the Shap method returns the importance of features, the Anchors method returns the minimum interpretable subset, and the Lime-based method returns the linear model.

[0085] It should be noted that this application does not limit the specific interpretability method adopted and the corresponding interpretation content, and can be universally applied to a variety of innate or acquired interpretability methods.

[0086] As an example, the explanation content of the explainability method can be the input attribute feature value, AI model, or comparison of explanation logic before and after model change, etc.

[0087] As an example, different interpretability methods may provide different interpretation accuracy and / or different numbers of feature attributes, etc. For example, different interpretability methods may provide different numbers of feature attributes corresponding to the inference results, and different feature value accuracy.

[0088] For example, the Sharp method can provide the number of features corresponding to the prediction result and whether it can provide the corresponding importance value and / or importance value.

[0089] Interpretable related information can be represented by the index-content-interpretation method shown in the table below, by establishing an index to indicate each category combination.

[0090] Table 1 Composition and corresponding relationship of model interpretability indicator information

[0091] S310: The receiving node sends first information to the model management node. In response, the model management node receives the first information. The first information indicates interpretable related information.

[0092] For the convenience of description, the interpretable related information indicated by the first information may be referred to as first interpretable related information.

[0093] In some possible implementations, the first information can be understood as the receiving node indicating to the model management node the required interpretability-related information. Alternatively, the receiving node sending the first information can be understood as the receiving node indicating the selected interpretability method.

[0094] In some possible implementations, when the receiving node sends the first information, the interpretable relevant information indicated by the first information may be determined by the second information.

[0095] As an example, the interpretable relevant information indicated by the first information can also be understood as the interpretable relevant information supported or recommended by the model management node.

[0096] In some possible implementations, the receiving node may feedback the selected interpretability method to the model management node according to the interpretability method indicated by the model management node; or, the receiving node may not feedback its own default interpretability method to the model management node according to the interpretability method indicated by the model management node.

[0097] As an example, when the receiving node specifies the interpretable relevant information supported or recommended by the model management node, the first information may be one of a plurality of interpretable relevant information, and the first information also includes an index of the first interpretable relevant information in the plurality of interpretable relevant information.

[0098] In the steps of this embodiment, the interpretable related information indicated by the first information may refer to the content indicated in Table 1, or may indicate interpretable related information that needs to be provided in addition to Table 1.

[0099] As an example, the first information may include index-based interpretability methods and parameter selections corresponding to different interpretability methods. For example, local interpretability methods such as Shap, Lime, and Anchors require information such as the selected local sample points or sample sets.

[0100] As an example, the first information may also directly feed back the first interpretable related information without indicating the first interpretable related information according to the index.

[0101] S320: The model management node performs AI model inference and performs explainability analysis on the inference results of the first AI model.

[0102] In some possible implementations, the inference result of the AI ​​model can be understood as the output result obtained by the model management node using the AI ​​model to train the predicted event.

[0103] As an example, for general AI application cases, such as using an AI model to infer and predict UE switching, the model management node can make the optimal switching decision through the first AI model, adaptively adjust the optimal parameters of UE switching, etc., and feed back the inference results of the first AI model to the UE, that is, the receiving node. Finally, the UE performs a switching operation based on the inference results (or output results) of the AI ​​model. This can reduce the delay of UE switching and improve the switching success rate.

[0104] It should be noted that the multi-node AI application interaction involved in the communication system of this application includes but is not limited to channel prediction, handover prediction, load balancing, and AI prediction information interaction between base stations. This application does not focus on the AI ​​application scenarios in the communication system. As long as it is a communication system that can realize the functional set of AI applications in the communication system, it should be included in the scope of protection of this application.

[0105] In some possible implementations, the model management node completes the reasoning of the AI ​​task according to the instructions returned by the receiving node, and performs model interpretability analysis based on the method agreed with the receiving node, or adds explainable explanations to the model's reasoning results.

[0106] In some possible implementations, explainability analysis can be understood as an explanatory analysis of the inference results of the AI ​​model using a specified explainable method.

[0107] S330: The model management node sends the inference result and the interpretable result of the first AI model to the receiving node. Correspondingly, the receiving node receives the inference result and the interpretable result of the first AI model.

[0108] In some possible implementations, the model management node performs an explainable analysis on the inference result of the first AI model to obtain an explainable result.

[0109] In this application, the interpretable result of the inference result of the AI ​​model can be the score of the interpretability test or the model obtained by the interpretability method, etc. The interpretable result of the inference result of the AI ​​model in this application can be referred to as the interpretability of the AI ​​model.

[0110] As an example, when the model management node's interpretable method based on the convention is Lime, the interpretable result is a simple linear model of local approximation, which is used to indicate that the features of some dimensions in the input data play a major role in the inference result; when the interpretability method is Shap, the interpretable result can be the feature weight value of one or more inputs for the inference result, which is used to indicate to the receiving node the impact of one or more input items on the inference result, thereby assisting the receiving node in confirming whether the inference result is logical; when the interpretability method is Anchors, the interpretable result is a set of features / if-then rule sets used to explain the model.

[0111] In the embodiment of the present application, after receiving the inference result of the AI ​​model, the receiving node needs to judge the credibility or rationality of the inference result to ensure the reliability of the AI ​​model applied to the communication network. The inference result of the first AI model obtained here can be understood as the relevant content information used for the subsequent judgment of its credibility and rationality.

[0112] In this embodiment, the model management node may send the inference result of the first AI model and the explainable result to the receiving node in different messages or in the same message.

[0113] In an embodiment of the present application, after the receiving node obtains an interpretable result of the inference result of the first AI model, optionally, it also includes S340.

[0114] S340: The receiving node sends fifth information to the model management node, indicating the rationality of the inference result of the first AI model. Accordingly, the model management node receives the fifth information.

[0115] As an example, the rationality of the model inference result can be understood as the receiving node accepting, recognizing or being reliable of the inference result, and the unreasonableness can be understood as not accepting, not recognizing or being unreliable.

[0116] In some possible implementations, when the receiving node sends the fifth information to the model management node, it indicates that the inference result is reasonable, and if it does not send it, it indicates that it is unreasonable; or, sending the fifth information indicates unreasonableness, and not sending the fifth information indicates reasonableness.

[0117] As an example, the receiving node determines whether to receive the relevant AI model output value or the operation indicated by the model strategy based on the rationality of the inference result.

[0118] In some possible implementations, if the receiving node receives the inference result, it means that the inference result is reasonable, then it will receive the relevant AI model output value or the operation indicated by the model strategy, and then proceed to perform subsequent tasks.

[0119] In some possible implementations, if the receiving node does not receive the inference result, it indicates the reason.

[0120] For example, the indication reason may be disapproval of the current interpretable result. For the Lime method, for example, the receiving node may indicate disapproval of the accuracy of the simple linear model and, therefore, disapproval of the inference result. For the Shap method, for example, the receiving node may indicate disapproval of the input item or its weight, i.e., disapproval of the influence of a specific input item on the inference result and, therefore, disapproval of the inference result. For the Anchors method, for example, the receiving node may indicate disapproval of the rule set or the sample points covered by the rule set.

[0121] For example, the reason for the indication could be a disapproval of the rationality of the inference result, meaning that the current interpretable result is not considered correct, or a need for a more precise interpretable result. For example, a more precise linear model may be required for the Lime method, or feature analysis results with more input parameters may be required for the Shap method. Alternatively, a general method may indicate the inclusion of more sample points for interpretability analysis.

[0122] This embodiment describes an interactive process in which the model management node, namely the AI ​​prediction and inference node, performs interpretable analysis. The prediction and inference node is responsible for providing both the AI ​​model's output and interpretable analysis. This allows the receiving node to determine the rationality and reliability of the AI ​​model output based on the interpretable analysis, allowing it to flexibly determine whether the inference result can be used for subsequent tasks. Furthermore, the AI ​​model's prediction and inference node provides the model's inference results and performs interpretable analysis, reducing the overhead of intermediate signaling interactions and effectively protecting the privacy of the AI ​​model.

[0123] In particular, this embodiment does not limit specific AI tasks and corresponding node instances. The corresponding interaction channel can select the connection channel of the corresponding network element, such as: Nx interface: the receiving node is gNB, the model management node is the core network side network element such as the computing management network element CMF, etc., and the base station requests the AI ​​model inference results or configures the scheduling strategy from the computing network element of the core network; Xn interface: the receiving node and the model management node are both gNB, such as the source station sends the prediction results based on the AI ​​model to the target station; F1 interface: the receiving node is gNB-DU, and the model management node is the centralized unit gNB-CU, which is responsible for the unified management of the AI ​​model; Uu interface: the receiving node is the terminal UE, the model management node is the base station gNB, and the terminal obtains the air interface information or configuration strategy predicted by the AI ​​model from the base station.

[0124] In a communication method provided in one embodiment of the present application, a management node performs reasoning on an AI task and sends the reasoning results and the input required for interpretability analysis to a receiving node. The receiving node then performs interpretability analysis on its own to facilitate subsequent tasks. An exemplary flowchart of the communication method of this embodiment is shown in FIG4 . The method may include S410 to S460.

[0125] S410: The model management node sends the inference result of the first AI model to the receiving node. Correspondingly, the receiving node receives the inference result of the first AI model.

[0126] In some possible implementations, the inference process of the AI ​​model has been completed before the model management node sends the inference result.

[0127] In this embodiment, this step can refer to the process of obtaining and sending the inference results of the AI ​​model in S320 and S330, and will not be repeated here.

[0128] S420: The receiving node sends first information to the model management node. Correspondingly, the model management node receives the first information. The first information indicates first interpretable related information.

[0129] As an example, the interpretable related information indicated by the first information may be interpretable related information supported by the receiving node.

[0130] In the steps of this embodiment, the interpretable related information indicated by the first information can be determined by referring to the content indicated in Table 1.

[0131] In some possible implementations, the interpretable related information may also include relevant inputs required for interpretable analysis, or in other words, input data required for interpretable analysis.

[0132] In some possible implementations, after receiving the inference result of the first AI model, the receiving node requests the model management to confirm the relevant input required for explanation. The specific content may include but is not limited to: the input amount corresponding to the relevant strategy, the inference result, etc.; the logical architecture of the model management node, etc.; the obtained parameters are used by the receiving node to make explainable judgments.

[0133] At step S430, the model management node sends third information to the receiving node. Accordingly, the receiving node receives the third information. The third information indicates information required to obtain an interpretable result of the inference result of the first AI model.

[0134] In some possible implementations, the receiving node may obtain the input data required for interpretability analysis or the input required for an interpretable method through third information, and perform interpretability analysis on the input data to obtain an interpretable result.

[0135] In some possible implementations, the model management node may indicate the input required by the interpretable method it supports and provide feedback to the receiving node.

[0136] For example, the model management node calculates the input required for the corresponding interpretable method based on the local AI model and provides feedback to the receiving node. For example, local interpretable methods such as Shap, Lime, and Anchors require information such as the selected local sample points / sets and the corresponding model output.

[0137] S440: The receiving node performs an explainability analysis on the inference result of the first AI model.

[0138] It can be understood that the receiving node performs interpretable analysis on the inference results or strategies based on the obtained model inference results and the input required by the interpretable method provided by the model management node.

[0139] In the embodiment of the present application, the interpretable analysis method is not limited and may be Shap, Anchors, Lime or other methods. The specific interpretable analysis steps are the same as those described in S320.

[0140] In the embodiment of the present application, after the receiving node obtains an interpretable result based on the inference result output by the model management node and the local interpretable analysis, optionally, S450 is also included.

[0141] S450: The receiving node sends fifth information to the model management node, indicating the rationality of the inference result of the first AI model. Accordingly, the model management node receives the fifth information.

[0142] In this embodiment, this step may refer to step S340 and will not be repeated here.

[0143] This embodiment provides an interactive process for interpretability analysis performed by a receiving node. The model management node provides auxiliary information for the receiving node to perform interpretability analysis on its own, so that the receiving node can flexibly perform interpretability analysis locally. The auxiliary information provided by the model management node can be reused in other subsequent interpretability analysis methods.

[0144] It should be noted that this embodiment does not limit the specific AI tasks and corresponding node instances. The corresponding interaction channel can select the connection channel of the corresponding network element. The details are as shown in the above embodiment and will not be repeated here.

[0145] In a communication method provided in one embodiment of the present application, a management node performs reasoning on an AI task and sends the reasoning results and the input required for interpretability analysis to a third node. The third node performs interpretability analysis and sends the interpretable results to the receiving node to facilitate the receiving node in performing subsequent tasks. An exemplary flow chart of the communication method of this embodiment is shown in Figure 5. The method may include S510 to S550.

[0146] S510: The model management node sends the inference result of the first AI model to the receiving node and the third node. Correspondingly, the receiving node and the third node receive the inference result of the first AI model.

[0147] It can be understood that the model management node can send the inference results of the first AI model to both the receiving node and the third node.

[0148] In this embodiment, the third node is a network node in the communication system that can be responsible for the interpretability analysis of the AI ​​model.

[0149] In this embodiment, this step may refer to S410 and will not be described again here.

[0150] S520: The third node sends first information to the model management node. Correspondingly, the model management node receives the first information. The first information indicates first interpretable related information.

[0151] In some possible implementations, the interpretable related information indicated by the first information may be interpretable related information supported by the third node.

[0152] In some possible implementations, the first information may be understood as the third node indicating to the model management node that the relevant information can be interpreted.

[0153] In this embodiment, the interpretable related information indicated in this step can be referred to S420 and will not be described in detail here.

[0154] In some possible implementations, it may also include: the third node sends fourth information to the model management node, where the fourth information is used to request information required to obtain the interpretability results of the inference results of the first AI model.

[0155] Alternatively, the third node may request the input data required for interpretability analysis through the fourth message. For example, the third node may send the fourth message when it determines an interpretable result.

[0156] At step S530, the model management node sends third information to the third node. Accordingly, the third node receives the third information. The third information indicates information required to obtain an interpretable result of the inference result of the first AI model.

[0157] In some possible implementations, the third information can be understood as the model management node indicating to the third node the input data required for interpretable analysis or the input required for the interpretable method.

[0158] In some possible implementations, the model management node may indicate the input required by the interpretable method it supports and provide feedback to the third node.

[0159] For example, the model management node calculates the input required for the corresponding interpretable method based on the local AI model and provides feedback to the third node. For example, local interpretable methods such as Shap, Lime, and Anchors require information such as the selected local sample points / sets and the corresponding model outputs.

[0160] S540: The third node performs an explainability analysis on the inference results of the first AI model.

[0161] In this embodiment, this step may refer to step S440 and will not be repeated here.

[0162] S550: The third node sends the interpretable result to the receiving node. Correspondingly, the receiving node receives the interpretable result.

[0163] In this embodiment, the third node determines whether to receive the operation indicated by the relevant output value or policy based on the output result of the model management node and the local interpretability analysis result, and feeds back the interpretability analysis result to the receiving node.

[0164] S560: The receiving node sends fifth information to the model management node, indicating the rationality of the inference result of the first AI model. Accordingly, the model management node receives the fifth information.

[0165] In some possible implementations, when the receiving node sends the fifth information to the model management node, it indicates that the inference result is reasonable, and if it does not send it, it indicates that it is unreasonable; or, sending the fifth information indicates unreasonableness, and not sending the fifth information indicates reasonableness.

[0166] In some possible implementations, the third node may directly send the indication information to the model management node, or the receiving node may send the indication information to the model management node.

[0167] In this embodiment, this step may refer to step S340 and will not be repeated here.

[0168] This embodiment provides an interactive process for introducing a third management node to perform interpretability analysis. The model management node provides auxiliary information, and the third node performs interpretability analysis in a unified manner. This makes the deployment and management of mobile network interpretability analysis more centralized and convenient to modify, while not adding additional computing load to the original nodes for analyzing model interpretability.

[0169] It should be noted that this embodiment does not limit the specific AI tasks and corresponding node instances. The corresponding interaction channel can select the connection channel of the corresponding network element. The specific model management node-receiving node-third management node instance can be: gNB-UE / gNB-model management network element, gNB(DU)-UE-gNB(CU), etc.

[0170] The embodiments of this application provide AI applications in communication networks that incorporate model interpretability analysis and interaction solutions, enabling receiving nodes to better judge the credibility of AI reasoning models. Furthermore, the three embodiments of this application differ in the deployment locations of the three types of interpretability analysis provided, such as deployment in the model management node, receiving node, or third-party management node.

[0171] FIG6 is a schematic diagram of the structure of a communication device according to an embodiment of the present application. As shown in FIG6 , the device 600 may include a processing module 601 and a communication module 602 .

[0172] As a first example, the apparatus 600 can be used to implement the communication method implemented by the model management node in any of the embodiments shown in Figures 3 to 5. For example, the processing module 601 is used to implement the processing-related steps performed by the model management node in any of the embodiments shown in Figures 3 to 5, and the communication module 602 is used to implement the sending and / or receiving steps performed by the model management node in any of the embodiments shown in Figures 3 to 5.

[0173] As a second example, the apparatus 600 may be used to implement the communication method implemented by a receiving node in any of the embodiments shown in Figures 3 to 5. For example, the processing module 601 is used to implement the processing-related steps performed by the receiving node in any of the embodiments shown in Figures 3 to 5, and the communication module 602 is used to implement the sending and / or receiving steps performed by the receiving node in any of the embodiments shown in Figures 3 to 5.

[0174] As a third example, the apparatus 600 may be used to implement the communication method implemented by a third node in any of the embodiments shown in Figures 3 to 5. For example, the processing module 601 is used to implement the processing-related steps performed by the third node in any of the embodiments shown in Figures 3 to 5, and the communication module 602 is used to implement the sending and / or receiving steps performed by the third node in any of the embodiments shown in Figures 3 to 5.

[0175] FIG7 is a schematic diagram of the structure of a communication device applicable to an embodiment of the present application. As shown in FIG7 , the communication device 700 includes a processor 711 , a memory 712 , and a transceiver 713 . The transceiver 713 includes a transmitter 7131 , a receiver 7132 , and an antenna 7133 .

[0176] The processor 711 , the memory 712 and the transceiver 713 communicate with each other through an internal connection path.

[0177] The receiver 7132 may be configured to receive transmission control information via the antenna 7133 , and the transmitter 7131 may be configured to send transmission feedback information to other communication nodes via the antenna 7133 .

[0178] Optionally, the apparatus 700 may further include a memory 712 for instructions executed by the processor 711 or input data required by the processor 711 to run instructions or data generated after the processor 711 runs instructions.

[0179] As an example, the processor 711 is used to implement the functions of the above-mentioned processing module 601, and the transceiver 713 is used to implement the functions of the above-mentioned communication module 602.

[0180] FIG8 is a schematic diagram of the structure of a communication device provided in another embodiment of the present application. As shown in FIG8 , the device 800 includes a processor 801 and a communication circuit 802. The processor 801 and the communication circuit 802 are coupled to each other. It is understood that the communication circuit 802 can be a transceiver or an input / output interface. Optionally, the device 800 may further include a memory 803 for storing instructions executed by the processor 801 or storing input data required by the processor 801 to run the instructions or storing data generated after the processor 801 runs the instructions. It is understood that the memory 803 can be located outside the processor 801, or inside the processor 801.

[0181] As an example, the processor 801 is used to implement the functions of the processing module 601 , and the communication circuit 802 is used to implement the functions of the communication module 602 .

[0182] The apparatus 800 may be a communication device or a chip used in a communication device. For example, the apparatus 800 may be a communication node or a chip used in a communication node. It is understood that when the apparatus 800 is a communication node, the communication circuit 802 may be a transceiver.

[0183] Some embodiments of the present application also provide a computer program product, which, when running on a processor, can implement the method implemented by the communication node in any of the above embodiments.

[0184] Some embodiments of the present application also provide a computer-readable storage medium, which includes computer instructions. When the computer instructions are executed on a processor, the method implemented by the communication node in any of the above embodiments can be implemented.

[0185] In some embodiments of the present application, a communication system is also provided, which can implement the method implemented by the communication node in any of the above embodiments.

[0186] It is understood that the processor in the embodiments of the present application can be the following devices or all or part of the circuits in the following devices for processing functions: a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor.

[0187] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, a register, a hard disk, a mobile hard disk, a CD-ROM or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a network device or a terminal device. Of course, the processor and the storage medium can also be present in a network device or a terminal device as discrete components.

[0188] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are performed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device, or other programmable device. The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video disk; or a semiconductor medium, such as a solid-state drive.

[0189] In the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0190] It is understood that the various numbers used in the embodiments of this application are merely for ease of description and are not intended to limit the scope of the embodiments of this application. The order of the sequence numbers of the above-mentioned processes does not necessarily imply a specific order of execution; the order of execution of the processes should be determined by their functions and inherent logic.

Claims

1. A communication method, characterized in that: Applied to a communication node, the method comprises: Obtaining the inference result of the first artificial intelligence AI model; Obtain an interpretable result of the inference result.

2. The method according to claim 1, characterized in that The method further comprises: Send first information, wherein the first information indicates first explainable related information, wherein the first explainable related information is explainable related information, and the explainable related information includes at least one of the following information: an explainability method, an output type, an explanation content of the explainability method, or an explanation accuracy of the explainability method, and the output type is the type of reasoning result of the AI ​​model.

3. The method according to claim 2, characterized in that The first information is one of a plurality of interpretable related information, wherein the first information includes an index of the first interpretable related information in the plurality of interpretable related information.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: Second information is received, wherein the second information indicates interpretable related information associated with each output type of at least one output type.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: receiving third information indicating information required to obtain an interpretable result of the inference result of the first AI model; Wherein, obtaining an interpretable result of the inference result includes: The explainable result is determined according to the third information.

6. The method according to claim 5, characterized in that The method further comprises: Send fourth information, where the fourth information is used to request information required to obtain an explainable result of the reasoning result of the first AI model.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: The interpretable result is transmitted.

8. The method according to any one of claims 1 to 4, characterized in that The obtaining of an interpretable result of the inferred result includes: The interpretable result is received.

9. The method according to any one of claims 1 to 8, characterized in that The method further comprises: Send fifth information, where the fifth information indicates the rationality of the reasoning result of the first AI model.

10. A communication device, characterized in that: The method comprises a functional module for implementing the method according to any one of claims 1 to 9.

11. A communication device, characterized in that: include: Memory and processor; The memory is used to store program instructions; The processor is configured to execute program instructions in the memory to implement the method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program code for computer execution, wherein the program code includes instructions for implementing the method according to any one of claims 1 to 9.

13. A computer program product, characterized in that The computer program product comprises instructions for implementing the communication method according to any one of claims 1 to 9.

Images