Ai model configuration method and apparatus, and storage medium

Abstract

The present disclosure provides an AI model configuration method and apparatus, and a storage medium. In the present disclosure, a first communication device sends first information to a second communication device, wherein the first information is used for indicating a use range of an AI model; and on the basis of the first information, the second communication device can guarantee the validity of the AI model within the use range indicated by the first information. Therefore, quantitative evaluation of a performance index, i.e., AI model generalization, can be implemented.

Description

AI model configuration methods, devices, and storage media Technical Field

[0001] This disclosure relates to the field of communications, and more particularly to a method and apparatus for configuring an AI model, and a storage medium. Background Technology

[0002] The 6th Generation Mobile Communication Technology (6G) era will see the emergence of various new intelligent application scenarios, and the artificial intelligence (AI) services inherent in 6G networks will provide strong support for the realization of these new intelligent application scenarios. For example, the 6G era will see various new intelligent application scenarios such as high-level network autonomy, intelligent and inclusive services for industry users, ultimate user service experience, and inherent network security. Different scenarios will have different requirements for the Quality of AI Service (QoAIS).

[0003] Summary of the Invention

[0004] To achieve a quantitative evaluation of the performance metric of model generalization, embodiments of this disclosure provide a method and apparatus for configuring an AI model, as well as a storage medium.

[0005] According to a first aspect of the present disclosure, a method for configuring an AI model is provided, applied to a first communication device, the method comprising:

[0006] Send first information to a second communication device. The first information is used to indicate the scope of use of the AI ​​model. The first information is also used by the second communication device to ensure the effectiveness of the AI ​​model within the scope of use.

[0007] According to a second aspect of the present disclosure, a method for configuring an AI model is provided, applied to a second communication device, the method comprising:

[0008] Receive first information sent by a first communication device, the first information being used to indicate the scope of use of the AI ​​model;

[0009] Based on the first information, the effectiveness of the AI ​​model within the scope of use is guaranteed.

[0010] According to a third aspect of the present disclosure, a first communication device is provided, comprising:

[0011] The transceiver module is configured to send first information to a second communication device. The first information is used to indicate the scope of use of the AI ​​model. The first information is also used by the second communication device to ensure the effectiveness of the AI ​​model within the scope of use.

[0012] According to a fourth aspect of the present disclosure, a second communication device is provided, comprising:

[0013] The transceiver module is configured to receive first information sent by the first communication device, the first information being used to indicate the scope of use of the AI ​​model;

[0014] The processing module is configured to ensure the effectiveness of the AI ​​model within the scope of use based on the first information.

[0015] According to a fifth aspect of the present disclosure, a first communication device is provided, comprising:

[0016] One or more processors;

[0017] The first communication device is used to execute the configuration method of the AI ​​model as described in the first aspect above.

[0018] According to a sixth aspect of the present disclosure, a second communication device is provided, comprising:

[0019] One or more processors;

[0020] The second communication device is used to execute the configuration method of the AI ​​model as described in the second aspect above.

[0021] According to a seventh aspect of the present disclosure, a communication system is provided, including a first communication device and a second communication device, wherein the first communication device is configured to implement the configuration method of the AI ​​model as described in the first aspect of the present disclosure, and the second communication device is configured to implement the configuration method of the AI ​​model as described in the second aspect above.

[0022] According to an eighth aspect of the present disclosure, a storage medium is provided that stores instructions that, when executed on a communication device, cause the communication device to perform the configuration method of the AI ​​model as described in the first or second aspect above.

[0023] According to a ninth aspect of the present disclosure, a program product is provided, the program product including a computer program, which, when executed by a communication device, causes the communication device to perform the configuration method of the AI ​​model as described in the first or second aspect above.

[0024] In this embodiment of the disclosure, a first communication device sends first information to a second communication device, and the second communication device receives the first information sent by the first communication device. The first information is used to indicate the scope of use of the AI ​​model, so that the second communication device can ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information, thereby achieving a quantitative evaluation of the performance index of the generalization of the AI ​​model.

[0025] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0026] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0027] Figure 1 is a schematic diagram of the architecture of a communication system according to an embodiment of the present disclosure.

[0028] Figure 2 is a schematic diagram illustrating the decomposition of a QoAIS metric into QoS metrics for each resource dimension according to an embodiment of the present disclosure.

[0029] Figure 3 is an interactive schematic diagram of an AI model configuration method according to an embodiment of the present disclosure.

[0030] Figure 4A is an interactive schematic diagram of an AI model configuration method according to an embodiment of the present disclosure.

[0031] Figure 4B is an interactive schematic diagram of an AI model configuration method according to an embodiment of the present disclosure.

[0032] Figure 4C is an interactive schematic diagram of an AI model configuration method according to an embodiment of the present disclosure.

[0033] Figure 5A is a flowchart illustrating a configuration method for an AI model according to an embodiment of the present disclosure.

[0034] Figure 5B is a flowchart illustrating a configuration method for an AI model according to an embodiment of the present disclosure.

[0035] Figure 5C is a flowchart illustrating a method for configuring an AI model according to an embodiment of the present disclosure.

[0036] Figure 5D is a flowchart illustrating a method for configuring an AI model according to an embodiment of the present disclosure.

[0037] Figure 5E is a flowchart illustrating a method for configuring an AI model according to an embodiment of the present disclosure.

[0038] Figure 6A is a schematic diagram of the structure of the first communication device proposed in an embodiment of this disclosure.

[0039] Figure 6B is a schematic diagram of the structure of the second communication device proposed in an embodiment of this disclosure.

[0040] Figure 7A is a schematic diagram of the structure of the communication device 7100 proposed in an embodiment of this disclosure.

[0041] Figure 7B is a schematic diagram of the structure of the chip 7200 proposed in the embodiments of this disclosure. Detailed Implementation

[0042] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0043] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of at least one associated listed item.

[0044] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various messages, these messages should not be limited to these terms. These terms are used only to distinguish messages of the same type from one another. For example, without departing from the scope of this disclosure, a first message may also be referred to as a second message, and similarly, a second message may also be referred to as a first message. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0045] This disclosure provides a method, apparatus, and storage medium for configuring an AI model.

[0046] In a first aspect, embodiments of this disclosure propose a method for configuring an AI model, applied to a first communication device, the method comprising:

[0047] Send first information to a second communication device. The first information is used to indicate the scope of use of the AI ​​model. The first information is also used by the second communication device to ensure the effectiveness of the AI ​​model within the scope of use.

[0048] In the above embodiments, by sending first information from the first communication device to the second communication device, the first information is used to indicate the scope of use of the AI ​​model, so that the second communication device can ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information, thereby achieving a quantitative evaluation of the performance index of the generalization of the AI ​​model.

[0049] In conjunction with some embodiments of the first aspect, in some embodiments, the scope of use of the AI ​​model includes at least one of the following:

[0050] The area of ​​application of the AI ​​model;

[0051] The usage time of the AI ​​model.

[0052] In the above embodiments, multiple optional forms of the scope of use of the AI ​​model are provided so that the scope of use of the AI ​​model can be flexibly configured from multiple aspects, thereby improving the flexibility of the generalization configuration process of the AI ​​model.

[0053] In conjunction with some embodiments of the first aspect, in some embodiments, the usage area of ​​the AI ​​model is determined based on at least one of the following:

[0054] Community signage information;

[0055] Geographic coordinate information;

[0056] Movement path information.

[0057] In the above embodiments, multiple optional information is provided for determining the usage area of ​​the AI ​​model, so that the usage area of ​​the AI ​​model can be determined based on multiple information, thereby improving the flexibility of the AI ​​model usage area determination process.

[0058] In conjunction with some embodiments of the first aspect, in some embodiments, the first information is further used to indicate the validity information of the scope of use, the validity information being used to indicate the proportion of the scope of use that the AI ​​model needs to ensure is valid within the scope of use indicated by the first information.

[0059] In the above embodiments, by using the first information to indicate the validity information of the scope of use, the flexibility and diversity of the content indicated by the first information are improved, and the second communication device can guarantee a portion of the scope of use indicated by the first information based on the indication of the first information, thereby improving the flexibility of the generalization configuration process of the AI ​​model.

[0060] In conjunction with some embodiments of the first aspect, in some embodiments, the validity information is used to indicate at least one of the following:

[0061] The AI ​​model must ensure that the effective usage time accounts for a proportion of the usage time indicated by the first information;

[0062] The AI ​​model must ensure that the effective usage area accounts for the proportion of the usage area indicated by the first information.

[0063] In the above embodiments, multiple optional information formats for validity information are provided to improve the flexibility and diversity of validity information.

[0064] In conjunction with some embodiments of the first aspect, in some embodiments, the first information is indicated by at least one of the following:

[0065] Performance requirements parameters;

[0066] Quality of Service (QoS) requirement information, which is used to determine the first information.

[0067] In the above embodiments, multiple optional information is provided for determining the first information, so that the first information can be determined based on multiple information, thereby improving the flexibility of the first information determination process.

[0068] In conjunction with some embodiments of the first aspect, in some embodiments, the first information is associated with an AI service, the AI ​​service including at least one of the following:

[0069] AI model training services;

[0070] AI model inference service;

[0071] AI model data services.

[0072] In the above embodiments, by associating AI services with first information, the corresponding AI model can be determined based on the first information; in addition, a variety of optional AI service types are provided so that the AI ​​services associated with the first information can be flexibly configured as needed, thereby improving the flexibility of the AI ​​model configuration process.

[0073] In conjunction with some embodiments of the first aspect, in some embodiments, the first information is used by the second communication device, based on the first information and the capability information of the third communication device, to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information through the computing power resources provided by the third communication device. The capability information is used to indicate the computing power resources provided by the third communication device, and / or, the capability information is used to indicate the scope of use for which the third communication device can provide a guarantee for the AI ​​model; or...

[0074] The first information is used by the second communication device to send resource request information to the third communication device. The resource request information is used to request the third communication device to provide computing power resources, and the computing power resources are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0075] In the above embodiments, the second communication device is provided with multiple optional interaction methods to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information, thereby improving the flexibility of the process of ensuring the effectiveness of the AI ​​model within the scope of use.

[0076] In conjunction with some embodiments of the first aspect, in some embodiments, the first information is used to assess the effectiveness of the second communication device AI model within the scope of use, including at least one of the following:

[0077] The first information is used by the second communication device to select a model training algorithm based on the first information;

[0078] The first information is used by the second communication device to collect first data for model training or model inference based on the first information, wherein the first data is data at a specific time and / or data at a specific location;

[0079] The first information is used by the second communication device to filter first data for model training or model inference, wherein the first data is data at a specific time and / or data at a specific location;

[0080] The first information is used by the second communication device to schedule computing resources based on the first information, and the computing resources are used to ensure the effectiveness of the AI ​​model within the scope of use.

[0081] In the above embodiments, multiple optional implementations are provided for the second communication device to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information, so as to improve the flexibility of the process of ensuring the effectiveness of the AI ​​model within the scope of use.

[0082] In conjunction with some embodiments of the first aspect, in some embodiments, the first communication device is a communication device using the AI ​​model, and / or, the first communication device is a communication device using the AI ​​service associated with the first information.

[0083] In the above embodiments, multiple optional device types of the first communication device are provided so that a suitable first communication device can be selected as needed, and the flexibility of the configuration process of the AI ​​model can be improved.

[0084] In conjunction with some embodiments of the first aspect, in some embodiments, the second communication device is used to provide at least one of the following functions:

[0085] The AI ​​model's control and management functions;

[0086] The training function of the AI ​​model;

[0087] The data functionality of the AI ​​model.

[0088] In the above embodiments, a variety of functions that the second communication device may provide are provided so that the functions that the second communication device can provide can be flexibly configured as needed.

[0089] Secondly, embodiments of this disclosure propose a method for configuring an AI model, applied to a second communication device, the method comprising:

[0090] Receive first information sent by a first communication device, the first information being used to indicate the scope of use of the AI ​​model;

[0091] Based on the first information, the effectiveness of the AI ​​model within the scope of use is guaranteed.

[0092] In the above embodiments, by receiving first information sent by the first communication device through the second communication device, the first information is used to indicate the scope of use of the AI ​​model, so that the second communication device can ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information, thereby achieving a quantitative evaluation of the performance index of the generalization of the AI ​​model.

[0093] In conjunction with some embodiments of the second aspect, in some embodiments, the scope of use of the AI ​​model includes at least one of the following:

[0094] The area of ​​application of the AI ​​model;

[0095] The usage time of the AI ​​model.

[0096] In conjunction with some embodiments of the second aspect, in some embodiments, the usage area of ​​the AI ​​model is determined based on at least one of the following:

[0097] Community signage information;

[0098] Geographic coordinate information;

[0099] Movement path information.

[0100] In conjunction with some embodiments of the second aspect, in some embodiments, the first information is further used to indicate the validity information of the scope of use, the validity information being used to indicate the proportion of the scope of use that the AI ​​model needs to ensure is valid within the scope of use indicated by the first information.

[0101] In conjunction with some embodiments of the second aspect, in some embodiments, the validity information is used to indicate at least one of the following:

[0102] The AI ​​model must ensure that the effective usage time accounts for a proportion of the usage time indicated by the first information;

[0103] The AI ​​model must ensure that the effective usage area accounts for the proportion of the usage area indicated by the first information.

[0104] In conjunction with some embodiments of the second aspect, in some embodiments, the first information is indicated by at least one of the following:

[0105] Performance requirements parameters;

[0106] Quality of Service (QoS) requirement information, which is used to determine the first information.

[0107] In conjunction with some embodiments of the second aspect, in some embodiments, the first information is associated with an AI service, the AI ​​service including at least one of the following:

[0108] AI model training services;

[0109] AI model inference service;

[0110] AI model data services.

[0111] In conjunction with some embodiments of the second aspect, in some embodiments, ensuring the effectiveness of the AI ​​model within the scope of use based on the first information includes:

[0112] Receive capability information sent by a third communication device, wherein the capability information is used to indicate the computing power resources provided by the third communication device, and / or, the capability information is used to indicate the scope of use that the third communication device can guarantee for the AI ​​model;

[0113] Based on the capability information and the first information, the computing power resources provided by the third communication device ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0114] In conjunction with some embodiments of the second aspect, in some embodiments, ensuring the effectiveness of the AI ​​model within the scope of use based on the first information includes:

[0115] Based on the first information, a resource request information is sent to a third communication device. The resource request information is used to request the third communication device to provide computing power resources, which are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0116] In conjunction with some embodiments of the second aspect, in some embodiments, the effectiveness of the AI ​​model within the scope of use is guaranteed by the second communication device based on its internal implementation, or the effectiveness of the AI ​​model within the scope of use is guaranteed by the second communication device based on effectiveness evaluation parameters obtained from external sources.

[0117] In the above embodiments, multiple optional implementations are provided for the second communication device to determine how to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information, so as to improve the flexibility of the process of ensuring the effectiveness of the AI ​​model within the scope of use.

[0118] In conjunction with some embodiments of the second aspect, in some embodiments, ensuring the effectiveness of the AI ​​model within the scope of use includes at least one of the following:

[0119] Select the model training algorithm based on the first information;

[0120] Based on the first information, first data is collected for model training or model inference, wherein the first data is data at a specific time and / or data at a specific location;

[0121] Based on the first information, first data for model training or model inference is selected, wherein the first data is data from a specific time and / or data from a specific location;

[0122] The computing resources are scheduled based on the first information, and the computing resources are used to ensure the effectiveness of the AI ​​model within the scope of use.

[0123] In conjunction with some embodiments of the second aspect, in some embodiments, the first communication device is a communication device that uses the AI ​​model, and / or the first communication device is a communication device that uses the AI ​​service associated with the first information.

[0124] In conjunction with some embodiments of the second aspect, in some embodiments, the second communication device is used to provide at least one of the following functions:

[0125] The AI ​​model's control and management functions;

[0126] The training function of the AI ​​model;

[0127] The data functionality of the AI ​​model.

[0128] Thirdly, embodiments of this disclosure provide a first communication device, comprising:

[0129] The transceiver module is configured to send first information to a second communication device. The first information is used to indicate the scope of use of the AI ​​model. The first information is also used by the second communication device to ensure the effectiveness of the AI ​​model within the scope of use.

[0130] Fourthly, embodiments of this disclosure provide a second communication device, comprising:

[0131] The transceiver module is configured to receive first information sent by the first communication device, the first information being used to indicate the scope of use of the AI ​​model;

[0132] The processing module is configured to ensure the effectiveness of the AI ​​model within the scope of use based on the first information.

[0133] Fifthly, embodiments of this disclosure provide a first communication device, comprising:

[0134] One or more processors;

[0135] The first communication device is used to execute the configuration method of the AI ​​model as described in the first aspect and any embodiment of the first aspect.

[0136] Sixthly, embodiments of this disclosure provide a second communication device, comprising:

[0137] One or more processors;

[0138] The second communication device is used to execute the configuration method of the AI ​​model as described in the second aspect and any embodiment of the second aspect.

[0139] In a seventh aspect, embodiments of this disclosure provide a communication system including a first communication device and a second communication device, wherein the first communication device is configured to implement the configuration method of the AI ​​model as described in the first aspect and any embodiment of the first aspect, and the second communication device is configured to implement the configuration method of the AI ​​model as described in the second aspect and any embodiment of the second aspect.

[0140] Eighthly, embodiments of this disclosure provide a storage medium storing instructions that, when executed on a communication device, cause the communication device to perform the configuration method of the AI ​​model as described in the first aspect and any embodiment of the first aspect, the second aspect and any embodiment of the second aspect.

[0141] In a ninth aspect, embodiments of this disclosure provide a program product comprising a computer program that, when executed by a communication device, causes the communication device to perform the configuration method of the AI ​​model as described in the first aspect and any embodiment of the first aspect, the second aspect and any embodiment of the second aspect.

[0142] In a tenth aspect, embodiments of this disclosure provide a program product that, when run on a communication device, causes the communication device to execute the configuration method of the AI ​​model as described in the first aspect and any embodiment of the first aspect, the second aspect and any embodiment of the second aspect, or, when the program product is executed by the communication device, causes the communication device to execute the configuration method of the AI ​​model as described in the first aspect and any embodiment of the first aspect, the second aspect and any embodiment of the second aspect.

[0143] In one aspect, embodiments of this disclosure provide a computer program that, when run on a computer, causes the computer to perform the configuration method of the AI ​​model as described in the first aspect and any embodiment of the first aspect, the second aspect and any embodiment of the second aspect.

[0144] In a twelfth aspect, embodiments of this disclosure provide a chip or chip system. The chip or chip system includes processing circuitry configured to execute the configuration method for an AI model as described in the first aspect and any embodiment thereof, the second aspect and any embodiment thereof.

[0145] It is understood that the aforementioned first communication device, second communication device, communication system, storage medium, program product, computer program, chip, or chip system are all used to execute the methods proposed in the embodiments of this disclosure. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0146] This disclosure provides an AI model configuration method, apparatus, and storage medium. In some embodiments, the terms "AI model configuration method" can be interchanged with "information processing method," "communication method," "AI model generalization evaluation method," and "AI model generalization quantitative evaluation method," etc. Similarly, the terms "AI model configuration apparatus" can be interchanged with "information processing apparatus," "communication apparatus," "AI model generalization evaluation apparatus," and "AI model generalization quantitative evaluation apparatus," etc., and the terms "information processing system" and "communication system," etc., can be interchanged.

[0147] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0148] In each of the disclosed embodiments, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0149] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0150] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0151] In the embodiments disclosed herein, "multiple" refers to two or more.

[0152] In some embodiments, the terms “at least one of”, “one or more”, “a plurality of”, “multiple”, etc., may be used interchangeably.

[0153] In some embodiments, the notation "at least one of A and B", "A and / or B", "A in one case, B in another", "in response to one case A, in response to another case B", etc., may include the following technical solutions depending on the situation: in some embodiments, A (execute A regardless of B); in some embodiments, B (execute B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed); in some embodiments, A and B (both A and B are executed). The same applies when there are more branches such as A, B, C, etc.

[0154] In some embodiments, the notation "A or B" may include the following technical solutions, depending on the situation: in some embodiments, A (execution of A regardless of B); in some embodiments, B (execution of B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed). The same applies when there are more branches such as A, B, C, etc.

[0155] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0156] In some embodiments, “including A,” “containing A,” “for indicating A,” and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

[0157] In some embodiments, the terms “in response to…”, “in response to determining…”, “in the case of…”, “when…”, “if…”, “if…”, etc., can be used interchangeably.

[0158] In some embodiments, the terms “greater than,” “greater than or equal to,” “not less than,” “more than,” “more than or equal to,” “not less than,” “higher than,” “higher than or equal to,” “not lower than,” and “above” can be used interchangeably, as can the terms “less than,” “less than or equal to,” “not greater than,” “less than,” “less than or equal to,” “not more than,” “lower than,” “lower than or equal to,” “not higher than,” and “below”.

[0159] In some embodiments, the apparatus and device may be interpreted as physical or virtual, and their names are not limited to the names recorded in the embodiments. In some cases, they may also be understood as "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "body", etc.

[0160] In some embodiments, "network" can be interpreted as devices included in the network, such as access network devices, core network devices, etc.

[0161] In some embodiments, "access network device (AN device)" may also be referred to as "radio access network device (RAN device)," "base station (BS)," "radio base station," or "fixed station." In some embodiments, it may also be understood as "node," "access point," "transmission point (TP)," "reception point (RP)," "transmission / reception point (TRP)," "panel," "antenna panel," "antenna array," "cell," "macro cell," "small cell," "femto cell," "pico cell," "sector," "cell group," "serving cell," "carrier," "component carrier," or "bandwidth part (BWP)."

[0162] In some embodiments, "terminal" or "terminal device" may be referred to as "user equipment (UE)," "user terminal," "mobile station (MS)," "mobile terminal (MT)," "subscriber station," "mobile unit," "subscriber unit," "wireless unit," "remote unit," "mobile device," "wireless device," "wireless communication device," "remote device," "mobile subscriber station," "access terminal," "mobile terminal," "wireless terminal," "remote terminal," "handset," "user agent," "mobile client," "client," etc.

[0163] In some embodiments, the acquisition of data, information, etc., may comply with the laws and regulations of the country where the location is situated.

[0164] In some embodiments, data, information, etc., may be obtained with the user's consent.

[0165] Furthermore, each element, each row, or each column in the table of this disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0166] Figure 1 is a schematic diagram of the architecture of a communication system according to an embodiment of the present disclosure. As shown in Figure 1, the communication system 100 includes a first communication device 101 and a second communication device 102.

[0167] In some embodiments, the first communication device 101 is a communication device using an AI model, and / or, the first communication device is a communication device using an AI service.

[0168] In some embodiments, the first communication device 101 includes at least one of user equipment (UE), access network device, and core network device.

[0169] In some embodiments, the name of the UE is not limited, and may be, for example, a terminal. The UE includes, but is not limited to, at least one of the following: mobile phone, wearable device, Internet of Things device, car with communication capabilities, smart car, tablet computer, computer with wireless transceiver capabilities, virtual reality (VR) terminal device, augmented reality (AR) terminal device, wireless terminal device in industrial control, wireless terminal device in self-driving, wireless terminal device in remote medical surgery, wireless terminal device in smart grid, wireless terminal device in transportation safety, wireless terminal device in smart city, and wireless terminal device in smart home.

[0170] In some embodiments, the access network device is, for example, a node or device that connects a UE to a wireless network. The access network device may include, but is not limited to, at least one of the following in a 5G communication system: evolved Node B (eNB), next-generation eNB (ng-eNB), next-generation Node B (gNB), node B (NB), home node B (HNB), home evolved node B (HeNB), radio backhaul device, radio network controller (RNC), base station controller (BSC), base transceiver station (BTS), base band unit (BBU), mobile switching center, base station in a 6G communication system, open RAN, cloud RAN, base station in other communication systems, and access node in a Wi-Fi system.

[0171] In some embodiments, the technical solutions of this disclosure can be applied to the Open RAN architecture. In this case, the interfaces between or within access network devices involved in the embodiments of this disclosure can be transformed into internal interfaces of Open RAN. The processes and information interactions between these internal interfaces can be implemented by software or programs.

[0172] In some embodiments, the access network device may be composed of a central unit (CU) and a distributed unit (DU). The CU may also be called a control unit. The CU-DU structure can separate the protocol layer of the access network device. Some of the protocol layer functions are centrally controlled by the CU, while the remaining part or all of the protocol layer functions are distributed in the DU and centrally controlled by the CU. However, this is not the only possibility.

[0173] In some embodiments, the core network equipment may be a single device comprising multiple network elements, or it may be multiple devices or a group of devices, each comprising all or part of the multiple network elements. Network elements may be virtual or physical. The core network may include, for example, at least one of the Evolved Packet Core (EPC), 5G Core Network (5GCN), and Next Generation Core (NGC).

[0174] In some embodiments, the core network equipment may include a first network element, such as an Access and Mobility Management Function (AMF).

[0175] In some embodiments, the first network element is used for user access management and mobility management, but is not limited thereto.

[0176] In some embodiments, the core network device may include a second network element, such as a Session Management Function (SMF).

[0177] In some embodiments, the second network element is used for session management of the control plane and user plane, but is not limited thereto.

[0178] In some embodiments, the core network device may include a third network element, such as a User Plane Function (UPF).

[0179] In some embodiments, the third network element is used for user plane data forwarding, traffic statistics, Quality of Service (QoS) management, etc., but is not limited to these.

[0180] In some embodiments, the core network device may include a fourth network element, such as a Policy Control Function (PCF).

[0181] In some embodiments, the fourth network element is used to implement user control policy management, including but not limited to QoS control, service access control, etc.

[0182] In some embodiments, the core network equipment may include a fifth network element, such as a unified data management function (UDM).

[0183] In some embodiments, the fifth network element is used to implement user subscription data management, roaming control, etc., but is not limited to these.

[0184] In some embodiments, the core network device may include a sixth network element, such as an Authentication Server Function (AUSF).

[0185] In some embodiments, the sixth network element is used to implement user authentication, but is not limited thereto.

[0186] In some embodiments, each of the above network elements can be independent of the core network equipment.

[0187] In some embodiments, each of the above network elements may be part of the core network equipment.

[0188] In some embodiments, the second communication device 102 may be used to provide at least one of the following functions: control and management functions of the AI ​​model, training functions of the AI ​​model, and data functions of the AI ​​model.

[0189] In some embodiments, the second communication device 102 includes at least one of a UE, an access network device, and a core network device.

[0190] For details regarding UE, access network equipment, and core network equipment, please refer to the above content; they will not be repeated here.

[0191] In many other possible implementations, the communication system 100 may also include a third communication device.

[0192] In some embodiments, the third communication device is a communication device capable of providing AI computing resources.

[0193] In some embodiments, the third communication device includes at least one of a UE, an access network device, and a core network device.

[0194] For details regarding UE, access network equipment, and core network equipment, please refer to the above content; they will not be repeated here.

[0195] It is understood that the communication system described in this disclosure is for the purpose of more clearly illustrating the technical solutions of this disclosure, and does not constitute a limitation on the technical solutions proposed in this disclosure. As those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions proposed in this disclosure are also applicable to similar technical problems.

[0196] The following embodiments of this disclosure can be applied to the communication system 100 shown in FIG1, or to some of the main bodies, but are not limited thereto. The main bodies shown in FIG1 are illustrative. The communication system may include all or some of the main bodies in FIG1, or may include other main bodies outside of FIG1. ​​The number and form of each main body are arbitrary. Each main body may be physical or virtual. The connection relationship between the main bodies is illustrative. The main bodies may not be connected or may be connected. The connection can be in any way, it can be a direct connection or an indirect connection, it can be a wired connection or a wireless connection.

[0197] The embodiments disclosed herein can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), and IEEE 802.20, Ultra-Wideband (UWB), Bluetooth (a registered trademark), Public Land Mobile Network (PLMN) networks, Device-to-Device (D2D) systems, Machine-to-Machine (M2M) systems, Internet of Things (IoT) systems, Vehicle-to-Everything (V2X) systems, systems utilizing other communication methods, and next-generation systems built upon them, etc. Furthermore, multiple systems can be combined (e.g., a combination of LTE or LTE-A with 5G).

[0198] For the various new intelligent application scenarios that will emerge in the 6G era (such as high-level network autonomy, intelligent accessibility for industry users, ultimate user business experience, and inherent network security), different scenarios will have different requirements for QoAIS. Therefore, a set of indicators is needed to express the user-level needs and the comprehensive effect of network orchestration control of various AI elements (including but not limited to algorithms, computing power, data, and connectivity) in a quantitative or hierarchical manner.

[0199] In 6G networks, in addition to traditional communication resources, various resource elements for AI service orchestration, such as distributed heterogeneous computing resources, storage resources, data resources, and AI algorithms, will be introduced. Among them, traditional communication resources mainly consider the latency, throughput (such as maximum bit rate (MBR) and guaranteed bit rate (GBR)) and connection-related performance indicators of communication services. Therefore, in 6G networks, it is necessary to comprehensively evaluate the service quality of network-inherent AI from multiple dimensions such as connectivity, computing power, algorithms, and data.

[0200] Meanwhile, with the implementation of "carbon neutrality" and "carbon peaking" policies, the increasing focus on data security and privacy in the global smart application industry, and the rising demand from users for network autonomy, the needs regarding security, privacy, autonomy, and resource consumption will gradually deepen, becoming new dimensions for evaluating service quality. The specific needs of different industries and scenarios in these new dimensions vary greatly, requiring quantitative or tiered assessments.

[0201] Therefore, the QoAIS indicator system needs to consider quantitative and hierarchical evaluations covering multiple aspects such as performance, overhead, security, privacy, and autonomy.

[0202] In some embodiments, AI services inherent in 6G networks can be categorized into AI data, AI training, AI inference, and AI verification, with each type of AI service requiring a QoAIS.

[0203] Taking AI training services as an example, their service quality can be evaluated using performance metrics across five dimensions: performance, overhead, security, privacy, and autonomy. The performance metrics for each dimension are shown in Table 1 below.

[0204] Table 1

[0205] In some embodiments, performance metric boundaries are used to indicate the upper and lower bounds of metrics used to evaluate the performance of a model. For example, performance metric boundaries can be used to indicate the range of performance metrics such as model error rate, precision, and recall, but are not limited to these.

[0206] In some embodiments, generalization refers to the ability of a model, after being trained, to be applied to new data and make accurate predictions.

[0207] In some embodiments, reusability is the ability of a model to continue to be used when the application scenario changes.

[0208] In some embodiments, robustness refers to the property that a model can still maintain certain performance characteristics when the input data is disturbed, attacked, or uncertain.

[0209] In some embodiments, interpretability refers to the degree to which a model supports an understanding of its internal mechanisms and its results.

[0210] In some embodiments, the consistency between the loss function and the optimization objective refers to the degree of consistency between the design of the loss function during model training and the optimization objective of the AI ​​use case. For example, whether the number of variables considered in the function fully covers the optimization objective metrics of the intelligent optimization scenario.

[0211] In some embodiments, autonomy refers to the requirements for the autonomous operation and human intervention portions of the workflow of an AI service (which can be any of AI data, AI training, AI verification, or AI inference services), reflecting the user's requirements for the degree of automation of the AI ​​service. Autonomy can be divided into three levels: fully autonomous (fully automated AI service with no human intervention required throughout), partially human-controlled (the workflow of the AI ​​service is automated in some stages and requires human assistance in others), and fully human-controlled (human participation is required in all stages of the AI ​​service workflow).

[0212] In some embodiments, QoAIS can be used as input to the network-native AI orchestration management system and control functions. The network-native AI management and orchestration system decomposes the top-level QoAIS and maps it to QoS requirements for data, algorithms, computing power, connectivity, and other aspects. Taking AI training services as an example, see Figure 2. Figure 2 is a schematic diagram of QoS indicators decomposed into various resource dimensions according to an embodiment of this disclosure. As shown in Figure 2, the top-level QoAIS can be decomposed into performance, security, privacy, overhead, and autonomy dimensions. Among them, performance indicators can be mapped to data resource dimensions, algorithm resource dimensions, computing power resource dimensions, and connection resource dimensions; security indicators can be mapped to data resource dimensions and algorithm resource dimensions; privacy indicators can be mapped to data resource dimensions and algorithm resource dimensions; overhead indicators can be mapped to storage resource dimensions, computing resource dimensions, transmission resource dimensions, and energy consumption resource dimensions; and autonomy indicators can be mapped to data resource dimensions and algorithm resource dimensions. In some embodiments, the data resource dimensions obtained by performance index mapping, the data resource dimensions obtained by security index mapping, the data resource dimensions obtained by privacy index mapping, the storage resource dimensions and computing resource dimensions obtained by overhead index mapping, and the data resource dimensions obtained by autonomy index mapping can be evaluated by data QoS indicators; the algorithm resource dimensions obtained by performance index mapping, the algorithm resource dimensions obtained by security index mapping, the algorithm resource dimensions obtained by privacy index mapping, the storage resource dimensions, computing resource dimensions and energy consumption resource dimensions obtained by overhead index mapping, and the algorithm resource dimensions obtained by autonomy index mapping can be evaluated by algorithm QoS indicators; the computing power resource dimensions obtained by performance index mapping, the computing resource dimensions and energy consumption resource dimensions obtained by overhead index mapping can be evaluated by computing power QoS indicators; and the connection resource dimensions obtained by performance index mapping, the transmission resource dimensions and energy consumption resource dimensions obtained by overhead index mapping can be evaluated by connection QoS indicators.

[0213] In some embodiments, QoS metrics for each resource dimension can be divided into metrics suitable for quantitative evaluation and metrics suitable for hierarchical evaluation. Taking the QoS metrics for each resource dimension obtained by the performance QoAIS mapping of AI training services as an example, the metrics suitable for quantitative evaluation and the metrics suitable for hierarchical evaluation obtained from these QoS metrics can be seen in Table 2 below:

[0214] Table 2

[0215] It is evident that there are currently no quantitative evaluation methods for some performance metrics of AI training services. Therefore, this disclosure proposes a scheme for quantitatively evaluating the generalization performance of AI models.

[0216] Figure 3 is an interactive schematic diagram illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 3, the embodiments of the present disclosure relate to an AI model configuration method, which includes:

[0217] Step S3101: The first communication device sends first information to the second communication device.

[0218] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0219] In some embodiments, the name of the first information is not limited, and it may be, for example, "first parameter information", "first indication information", "model parameter information", "model usage scope information", etc.

[0220] In some embodiments, the names of information, etc., are not limited to the names described in the embodiments. Terms such as "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "domain", "field", "symbol", "symbol", "codebook", "codeword", "codepoint", "bit", "data", "program", and "chip" can be used interchangeably.

[0221] In some embodiments, the scope of use of the AI ​​model includes both time and location dimensions; that is, the scope of use of the AI ​​model includes the time of use and / or the area of ​​use. Specifically, the first information can be used to indicate the time of use and / or the area of ​​use of the AI ​​model, so that the effective coverage of the AI ​​model within a specific time range and / or area can be guaranteed based on the first information, ensuring the effectiveness of the AI ​​model within that specific time range and / or area, thereby achieving generalization assurance for the AI ​​model.

[0222] In some embodiments, the area where the AI ​​model can be used can be determined based on at least one of cell identification information, geographic coordinate information, and movement path information.

[0223] In some embodiments, the first information is further used to indicate the validity information of the scope of use. The validity information is used to indicate the proportion of the scope of use in which the AI ​​model must ensure its validity within the scope of use indicated by the first information. In other words, the validity information is used to indicate the proportion of the scope of use in which the AI ​​model must ensure its validity within the scope of use indicated by the first information.

[0224] In some embodiments, corresponding to the application scope including time and location dimensions, the validity information includes the validity percentage information of the time dimension and the validity percentage information of the location dimension.

[0225] In other words, validity information can be used to indicate the proportion of the usage time that the AI ​​model needs to ensure is valid within the usage time indicated by the first information, and / or, validity information can be used to indicate the proportion of the usage area that the AI ​​model needs to ensure is valid within the usage area indicated by the first information; or, validity information can be used to indicate the proportion of the usage time that the AI ​​model needs to ensure is valid within the usage area indicated by the first information, and / or, validity information can be used to indicate the proportion of the usage area that the AI ​​model needs to ensure is valid within the usage area indicated by the first information.

[0226] In some embodiments, the effective proportion information in the time dimension indicates the time range within which the AI ​​model needs to be guaranteed to be available, within the proportion indicated by the effective proportion information. The effective proportion information in the time dimension indicates the effective coverage proportion that needs to be guaranteed within a specific time range, thereby ensuring the generalization ability of the AI ​​model.

[0227] In some embodiments, the effective coverage percentage information in the location dimension indicates the area within which the AI ​​model is used, ensuring that the AI ​​model is usable within the location area indicated by the effective coverage percentage information. This effective coverage percentage information in the location dimension indicates the effective coverage percentage that needs to be guaranteed within a specific area, thereby ensuring the generalization ability of the AI ​​model.

[0228] In some embodiments, the first information can be indicated by performance requirement parameters. Optionally, the first communication device can directly indicate the first information as a performance requirement parameter to the second communication device.

[0229] In some embodiments, the first information can be indicated by Quality of Service (QoS) requirement information, for example, by an identifier of the QoS requirement. Optionally, the first communication device can indicate the identifier of the QoS requirement to the second communication device, and the second communication device determines the first information based on the identifier of the QoS requirement indicated by the first communication device.

[0230] In some embodiments, the first information can be indicated by performance requirement parameters and QoS requirement information. For example, a portion of the first information can be indicated by performance requirement parameters, and another portion by QoS requirement information, so that the complete first information can be indicated by both performance requirement parameters and QoS requirement information.

[0231] In some embodiments, the first information has a default value, so that if the first information is not specified or indicated, the first information can be determined according to the default value.

[0232] In some embodiments, if the first information does not have a default value, then the value of the first information needs to be specified or indicated. Optionally, if the first information does not have a default value, and no value of the first information is specified or indicated, then it is considered that there is no corresponding performance requirement or that the performance requirement is invalid.

[0233] In some embodiments, the first information may be associated with AI services, which may include, but are not limited to, at least one of AI model training services, AI model inference services, and AI model data services.

[0234] In some embodiments, if the first information is associated with an AI service, the first communication device can send a service request for the AI ​​service to the second communication device. The service request may directly include the first information in order to send the first information to the second communication device.

[0235] In some embodiments, if the first information is associated with an AI service, the first communication device can send a service request for the AI ​​service to the second communication device. The service request includes AI service information, and the AI ​​service is associated with the first information by default. The second communication device can determine the associated first information based on the AI ​​service information in order to send the first information to the second communication device.

[0236] In some embodiments, the second communication device receives the first information sent by the first communication device.

[0237] In some embodiments, the first communication device is a communication device that uses an AI model, that is, the first communication device can be an AI model consumer; and / or, the first communication device is a communication device that uses an AI service associated with the first information, that is, the first communication device can be an AI service consumer.

[0238] In some embodiments, the name of the first communication device is not limited, and it may be, for example, "first node".

[0239] In some embodiments, the second communication device is used to provide at least one of the following functions: AI model control and management function, AI model training function, and AI model data function, but is not limited thereto. That is, the second communication device can be at least one of the following functions: AI control and management node, AI model training node, and AI data node, but is not limited thereto.

[0240] In some embodiments, the name of the second communication device is not limited, and it may be, for example, "second node".

[0241] In step S3102, the second communication device, based on the first information, ensures the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0242] In some embodiments, the second communication device receives capability information sent by the third communication device, so as to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information by using the computing resources provided by the third communication device based on the capability information and the first information.

[0243] In some embodiments, the third communication device is a communication device capable of providing computing power, that is, the third communication device provides nodes for AI computing power resources.

[0244] In some embodiments, the name of the third communication device is not limited, and it may be, for example, a "third node" or an "AI computing resource provider node".

[0245] In some embodiments, capability information is used to indicate the computing resources provided by the third communication device, and / or, capability information is used to indicate the scope of use by which the third communication device can provide guarantees for the AI ​​model.

[0246] In some embodiments, where the capability information is used to indicate the scope of use in which the third communication device can provide assurance for the AI ​​model, the capability information is used to indicate the duration of use in which the third communication device can provide assurance for the AI ​​model, and / or, the capability information is used to indicate the area of ​​use in which the third communication device can provide assurance for the AI ​​model.

[0247] In some embodiments, the second communication device sends resource request information to the third communication device based on the first information, and the third communication device receives the resource request information sent by the second communication device to provide computing resources to the second communication device based on the resource request information.

[0248] In some embodiments, the resource request information is used to request a third communication device to provide computing resources, which are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0249] In some embodiments, the first information can be directly used as the resource request information; or, the computing power resources required to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information can be determined based on the first information, thereby requesting the required computing power resources through the resource request information.

[0250] In some embodiments, if the second communication device is unable to provide AI computing resources or the AI ​​computing resources provided by the second communication device are insufficient to guarantee the effectiveness of the AI ​​model within the scope of use indicated by the first information, the second communication device may receive capability information sent by the third communication device, and based on the capability information and the first information, guarantee the effectiveness of the AI ​​model within the scope of use indicated by the first information through the computing resources provided by the third communication device; or, the second communication device may send resource request information to the third communication device, and guarantee the effectiveness of the AI ​​model within the scope of use indicated by the first information through the computing resources provided by the third communication device.

[0251] In some embodiments, the second communication device, based on the first information, configures sufficient computing resources through internal resource scheduling to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0252] In some embodiments, if the second communication device has computing power resources or the computing power resources provided by the second communication device can meet the determination of the first communication device, then the second communication device can configure sufficient computing power resources based on the first information through internal resource scheduling configuration to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0253] In some embodiments, the second communication device ensures the effectiveness of the AI ​​model within the scope of use indicated by the first information. The effectiveness may be determined by the second communication device based on its internal implementation, or it may be determined by the second communication device based on effectiveness evaluation parameters obtained from external sources.

[0254] In some embodiments, the second communication device may also assess whether its own computing resources can guarantee the effectiveness of the AI ​​model within its scope of use based on internally implemented or externally obtained evaluation parameters; or, the second communication device may also assess the computing resources required to guarantee the effectiveness of the AI ​​model within its scope of use based on internally implemented or externally obtained evaluation parameters.

[0255] In some embodiments, the second communication device can ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information in the following ways:

[0256] Select the model training algorithm based on the first piece of information;

[0257] First data is collected based on first information for model training or model inference, wherein the first data is data at a specific time and / or data at a specific location;

[0258] First data for model training or model inference is selected based on first information, where the first data is data from a specific time and / or data from a specific location;

[0259] The computing resources are allocated based on the first piece of information, and these resources are used to ensure the effectiveness of the AI ​​model within the scope indicated by the first piece of information.

[0260] It should be noted that the above are only a few exemplary implementation methods and do not constitute a limitation on the embodiments of this disclosure. In more possible implementation methods, the effectiveness of the AI ​​model within the scope of use indicated by the first information can also be achieved in other ways.

[0261] In some embodiments, the accuracy, completeness, and consistency of data have a direct impact on the generalization of AI models. Therefore, the effectiveness of AI models within the scope of use indicated by the first information can be guaranteed by improving data quality (including but not limited to improving data accuracy, data completeness, and data consistency).

[0262] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by increasing data diversity. By increasing data diversity, the training data can cover as many possible input scenarios as possible, thereby minimizing the risk of the AI ​​model overfitting to certain input scenarios.

[0263] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by increasing the amount of data. By increasing the amount of data, sufficient training data can help the AI ​​model learn a wider range of features, thereby improving the generalization ability of the AI ​​model.

[0264] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by optimizing the data distribution. By optimizing the data distribution, the distribution of the training data becomes more similar to the data distribution of the actual application, ensuring that the AI ​​model can generalize to new data.

[0265] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by optimizing model selection. By optimizing model selection to choose an appropriate model architecture, it is possible to avoid overfitting due to an overly complex model, as well as to avoid an overly simple model that fails to capture the complexity of the data.

[0266] In some embodiments, continuous learning can be used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the initial information. Continuous learning allows for the collection and updating of new data after model deployment, ensuring that the AI ​​model can adapt to changes in data distribution and thus improving its generalization ability.

[0267] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by optimizing feature extraction. Optimizing feature extraction enables effective feature selection and transformation, thereby improving the generalization ability of the AI ​​model.

[0268] In some embodiments, ensemble learning can be used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information. Ensemble learning combines the predictions of multiple models to improve the generalization ability of the AI ​​model.

[0269] In some embodiments, the effectiveness of the AI ​​model within the scope indicated by the first information can be guaranteed by employing regularization techniques in the AI ​​model. For example, L1 regularization, L2 regularization, dropout, and other regularization techniques can be used, but are not limited to these. By employing regularization techniques, model complexity can be reduced and model generalization can be improved.

[0270] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by optimizing the training method of the AI ​​model. For example, gradient descent, optimizer selection, and other methods can be used to optimize the model training method, but this is not limited to these. By optimizing the model training method, the training process and the final training performance of the model can be optimized, thereby improving the generalization ability of the model.

[0271] In some embodiments, the effectiveness of the AI ​​model within the scope indicated by the first information can be guaranteed by optimizing hyperparameter settings. For example, the settings of hyperparameters such as learning rate, batch size, and number of iterations can be optimized. Hyperparameter settings affect the training effect and generalization ability of the model; by optimizing hyperparameter settings, the training effect and generalization ability of the model can be improved.

[0272] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by optimizing the hardware environment. The hardware environment for training and deploying the model may affect the model's generalization ability, and optimizing the hardware environment can improve the model's generalization ability.

[0273] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information can be guaranteed by optimizing the software environment. The software environment for training and deploying the model may affect the model's generalization ability, and optimizing the software environment can improve the model's generalization ability.

[0274] It should be noted that the above-mentioned methods for ensuring the effectiveness of the AI ​​model within the scope of use indicated by the first information can be used in combination, and this disclosure does not limit this.

[0275] In some embodiments, “get,” “obtain,” “receive,” “transmit,” “bidirectional transmission,” and “send and / or receive” can be used interchangeably and can be interpreted as receiving from other entities, obtaining from protocols, obtaining from higher layers, obtaining through self-processing, or autonomous implementation, among other meanings.

[0276] In some embodiments, terms such as “send,” “transmit,” “report,” “distribute,” “transfer,” “bidirectional transmission,” “send and / or receive” can be used interchangeably.

[0277] In some embodiments, terms such as "certain," "preset," "default," "set," "indicated," "a certain," "any," and "first" can be used interchangeably. "Certain A," "preset A," "default A," "set A," "indicated A," "a certain A," "any A," and "first A" can be interpreted as A pre-defined in a protocol or the like, or as A obtained through setting, configuration, or instruction, or as specific A, a certain A, any A, or first A, but are not limited thereto.

[0278] In some embodiments, the determination or judgment can be made by a value represented by 1 bit (0 or 1), or by a true or false value (boolean), or by a comparison of numerical values ​​(e.g., a comparison with a predetermined value), but is not limited thereto.

[0279] In some embodiments, "not expecting to receive" can be interpreted as not receiving on time domain resources and / or frequency domain resources, or as not performing subsequent processing on the data after receiving it; "not expecting to send" can be interpreted as not sending, or as sending but not expecting the receiver to respond to the sent content.

[0280] The communication method involved in the embodiments of this disclosure may include at least one of steps S3101 to S3102. For example, step S3101 may be implemented as a standalone embodiment, step S3102 may be implemented as a standalone embodiment, and step S3101+S3102 may be implemented as a standalone embodiment, but is not limited thereto.

[0281] In some embodiments, step S3101 is optional and may be omitted or replaced in different embodiments.

[0282] In some embodiments, step S3102 is optional and may be omitted or replaced in different embodiments.

[0283] In some embodiments, other optional implementations may be described before or after the specification corresponding to FIG3.

[0284] Figure 4A is an interactive schematic diagram illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 4A, the embodiments of the present disclosure relate to an AI model configuration method, which includes:

[0285] Step S4101: The first communication device sends first information to the second communication device.

[0286] The optional implementation of step S4101 can be found in the optional implementation of step S3101 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0287] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0288] In some embodiments, the scope of use of the AI ​​model includes the time and / or the area where the AI ​​model is used.

[0289] In some embodiments, the area where the AI ​​model can be used can be determined based on at least one of cell identification information, geographic coordinate information, and movement path information.

[0290] In some embodiments, the first information is further used to indicate validity information of the scope of use, which indicates the proportion of the scope of use in which the AI ​​model must ensure its validity within the scope of use indicated by the first information.

[0291] In some embodiments, validity information may be used to indicate the proportion of the usage time that the AI ​​model needs to guarantee is valid within the usage time indicated by the first information, and / or, validity information may be used to indicate the proportion of the usage area that the AI ​​model needs to guarantee is valid within the usage area indicated by the first information.

[0292] In some embodiments, the first information may be indicated by performance requirement parameters, and / or the first information may be indicated by QoS requirement information.

[0293] In some embodiments, the first information may be associated with AI services, which may include, but are not limited to, at least one of AI model training services, AI model inference services, and AI model data services.

[0294] In some embodiments, the second communication device receives the first information sent by the first communication device.

[0295] In some embodiments, the first communication device is a communication device that uses an AI model, and / or the first communication device is a communication device that uses an AI service associated with the first information.

[0296] In some embodiments, the second communication device is used to provide at least one of the following functions: control and management functions for the AI ​​model, training functions for the AI ​​model, and data functions for the AI ​​model.

[0297] In some embodiments, the second communication device is unable to provide AI computing resources, or the AI ​​computing resources provided by the second communication device are insufficient to guarantee the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0298] In step S4102, the third communication device sends capability information to the second communication device.

[0299] The optional implementation of step S4102 can be found in the optional implementation of step S3102 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0300] In some embodiments, the third communication device is a communication device capable of providing computing power.

[0301] In some embodiments, the second communication device receives capability information sent by the third communication device.

[0302] In some embodiments, capability information is used to indicate the computing resources provided by the third communication device, and / or, capability information is used to indicate the scope of use by which the third communication device can provide guarantees for the AI ​​model.

[0303] In some embodiments, capability information is used to indicate the scope of use in which the third communication device can provide protection for the AI ​​model, the duration of use in which the third communication device can provide protection for the AI ​​model, and / or, the area of ​​use in which the third communication device can provide protection for the AI ​​model.

[0304] In step S4103, the third communication device provides computing resources to the second communication device.

[0305] The optional implementation of step S4103 can be found in the optional implementation of step S3102 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0306] In step S4104, the second communication device, based on the first information and capability information, uses the computing power resources provided by the third communication device to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0307] The optional implementation of step S4103 can be found in the optional implementation of step S3102 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0308] In some embodiments, the model training algorithm is selected based on the first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0309] In some embodiments, first data for model training or model inference is collected based on first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information. The first data is data at a specific time and / or data at a specific location.

[0310] In some embodiments, first data for model training or model inference is filtered according to first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information. The first data is data from a specific time and / or data from a specific location.

[0311] In some embodiments, computing resources are scheduled based on first information, and the computing resources are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0312] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by improving data quality (including but not limited to improving data accuracy, data integrity, and data consistency).

[0313] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by increasing data diversity.

[0314] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by increasing the amount of data.

[0315] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing the data distribution.

[0316] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing model selection.

[0317] In some embodiments, continuous learning is used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0318] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing feature extraction.

[0319] In some embodiments, ensemble learning is used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0320] In some embodiments, regularization techniques are employed in the AI ​​model to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0321] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the training method of the AI ​​model.

[0322] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing hyperparameter settings.

[0323] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the hardware environment for training and deploying the model.

[0324] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the software environment for training and deploying the model.

[0325] The communication method involved in the embodiments of this disclosure may include at least one of steps S4101 to S4104. For example, step S4101 can be implemented as an independent embodiment, step S4104 can be implemented as an independent embodiment, step S4101+S4102 can be implemented as an independent embodiment, step S4101+S4103 can be implemented as an independent embodiment, step S4101+S4104 can be implemented as an independent embodiment, step S4102+S4104 can be implemented as an independent embodiment, step S4103+S4104 can be implemented as an independent embodiment, step S4101+S4102+S4103 can be implemented as an independent embodiment, step S4101+S4102+S4104 can be implemented as an independent embodiment, step S4101+S4103+S4104 can be implemented as an independent embodiment, but not limited thereto.

[0326] In some embodiments, steps S4101 and S4102 can be performed in a different order or simultaneously, and steps S4101 and S4103 can be performed in a different order or simultaneously.

[0327] In some embodiments, steps S4102, S4103, and S4104 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0328] In some embodiments, steps S4101, S4102, and S4103 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0329] Figure 4B is an interactive schematic diagram illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 4B, the embodiments of the present disclosure relate to an AI model configuration method, which includes:

[0330] Step S4201: The first communication device sends first information to the second communication device.

[0331] The optional implementation of step S4201 can be found in the optional implementation of step S3101 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0332] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0333] In some embodiments, the scope of use of the AI ​​model includes the time and / or the area where the AI ​​model is used.

[0334] In some embodiments, the area where the AI ​​model can be used can be determined based on at least one of cell identification information, geographic coordinate information, and movement path information.

[0335] In some embodiments, the first information is further used to indicate validity information of the scope of use, which indicates the proportion of the scope of use in which the AI ​​model must ensure its validity within the scope of use indicated by the first information.

[0336] In some embodiments, validity information may be used to indicate the proportion of the usage time that the AI ​​model needs to guarantee is valid within the usage time indicated by the first information, and / or, validity information may be used to indicate the proportion of the usage area that the AI ​​model needs to guarantee is valid within the usage area indicated by the first information.

[0337] In some embodiments, the first information may be indicated by performance requirement parameters, and / or the first information may be indicated by QoS requirement information.

[0338] In some embodiments, the first information may be associated with AI services, which may include, but are not limited to, at least one of AI model training services, AI model inference services, and AI model data services.

[0339] In some embodiments, the second communication device receives the first information sent by the first communication device.

[0340] In some embodiments, the first communication device is a communication device that uses an AI model, and / or the first communication device is a communication device that uses an AI service associated with the first information.

[0341] In some embodiments, the second communication device is used to provide at least one of the following functions: control and management functions for the AI ​​model, training functions for the AI ​​model, and data functions for the AI ​​model.

[0342] In some embodiments, the second communication device is unable to provide AI computing resources, or the AI ​​computing resources provided by the second communication device are insufficient to guarantee the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0343] In step S4202, the second communication device sends a resource request message to the third communication device based on the first information.

[0344] The optional implementation of step S4202 can be found in the optional implementation of step S3102 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0345] In some embodiments, the third communication device receives resource request information sent by the second communication device.

[0346] In some embodiments, the third communication device is a communication device capable of providing computing power.

[0347] In some embodiments, the resource request information is used to request a third communication device to provide computing resources, which are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0348] In step S4203, the third communication device provides computing resources to the second communication device based on the resource request information.

[0349] The optional implementation of step S4203 can be found in the optional implementation of step S3102 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0350] In step S4204, the second communication device uses the computing power resources provided by the third communication device to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0351] The optional implementation of step S4204 can be found in the optional implementation of step S3102 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0352] In some embodiments, the model training algorithm is selected based on the first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0353] In some embodiments, first data for model training or model inference is collected based on first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information. The first data is data at a specific time and / or data at a specific location.

[0354] In some embodiments, first data for model training or model inference is filtered according to first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information. The first data is data from a specific time and / or data from a specific location.

[0355] In some embodiments, computing resources are scheduled based on first information, and the computing resources are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0356] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by improving data quality (including but not limited to improving data accuracy, data integrity, and data consistency).

[0357] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by increasing data diversity.

[0358] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by increasing the amount of data.

[0359] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing the data distribution.

[0360] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing model selection.

[0361] In some embodiments, continuous learning is used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0362] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing feature extraction.

[0363] In some embodiments, ensemble learning is used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0364] In some embodiments, regularization techniques are employed in the AI ​​model to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0365] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the training method of the AI ​​model.

[0366] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing hyperparameter settings.

[0367] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the hardware environment for training and deploying the model.

[0368] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the software environment for training and deploying the model.

[0369] The communication method involved in the embodiments of this disclosure may include at least one of steps S4201 to S4204. For example, step S4201 may be implemented as an independent embodiment, step S4204 may be implemented as an independent embodiment, step S4201+S4202 may be implemented as an independent embodiment, step S4201+S4203 may be implemented as an independent embodiment, step S4201+S4204 may be implemented as an independent embodiment, step S4202+S4204 may be implemented as an independent embodiment, step S4203+S4204 may be implemented as an independent embodiment, step S4201+S4202+S4203 may be implemented as an independent embodiment, and step S4202+S4203+S4204 may be implemented as an independent embodiment, but is not limited thereto.

[0370] In some embodiments, steps S4202, S4203, and S4204 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0371] In some embodiments, steps S4201, S4202, and S4203 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0372] Figure 4C is an interactive schematic diagram illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 4C, the embodiments of the present disclosure relate to an AI model configuration method, which includes:

[0373] Step S4301: The first communication device sends first information to the second communication device.

[0374] The optional implementation of step S4301 can be found in the optional implementation of step S3101 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0375] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0376] In some embodiments, the scope of use of the AI ​​model includes the time and / or the area where the AI ​​model is used.

[0377] In some embodiments, the area where the AI ​​model can be used can be determined based on at least one of cell identification information, geographic coordinate information, and movement path information.

[0378] In some embodiments, the first information is further used to indicate validity information of the scope of use, which indicates the proportion of the scope of use in which the AI ​​model must ensure its validity within the scope of use indicated by the first information.

[0379] In some embodiments, validity information may be used to indicate the proportion of the usage time that the AI ​​model needs to guarantee is valid within the usage time indicated by the first information, and / or, validity information may be used to indicate the proportion of the usage area that the AI ​​model needs to guarantee is valid within the usage area indicated by the first information.

[0380] In some embodiments, the first information may be indicated by performance requirement parameters, and / or the first information may be indicated by QoS requirement information.

[0381] In some embodiments, the first information may be associated with AI services, which may include, but are not limited to, at least one of AI model training services, AI model inference services, and AI model data services.

[0382] In some embodiments, the second communication device receives the first information sent by the first communication device.

[0383] In some embodiments, the first communication device is a communication device that uses an AI model, and / or the first communication device is a communication device that uses an AI service associated with the first information.

[0384] In some embodiments, the second communication device is used to provide at least one of the following functions: control and management functions for the AI ​​model, training functions for the AI ​​model, and data functions for the AI ​​model.

[0385] In some embodiments, the second communication device has computing power resources, or the computing power resources provided by the second communication device can meet the requirements of the first communication device.

[0386] In step S4302, the second communication device, based on the first information, ensures the effectiveness of the AI ​​model within the scope of use indicated by the first information through internal resource scheduling configuration.

[0387] The optional implementation of step S4302 can be found in the optional implementation of step S3102 in Figure 3 and other related parts in the embodiments involved in Figure 3, which will not be repeated here.

[0388] In some embodiments, the model training algorithm is selected based on the first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0389] In some embodiments, first data for model training or model inference is collected based on first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information. The first data is data at a specific time and / or data at a specific location.

[0390] In some embodiments, first data for model training or model inference is filtered according to first information to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information. The first data is data from a specific time and / or data from a specific location.

[0391] In some embodiments, computing resources are scheduled based on first information, and the computing resources are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0392] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by improving data quality (including but not limited to improving data accuracy, data integrity, and data consistency).

[0393] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by increasing data diversity.

[0394] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by increasing the amount of data.

[0395] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing the data distribution.

[0396] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing model selection.

[0397] In some embodiments, continuous learning is used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0398] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing feature extraction.

[0399] In some embodiments, ensemble learning is used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0400] In some embodiments, regularization techniques are employed in the AI ​​model to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0401] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the training method of the AI ​​model.

[0402] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by optimizing hyperparameter settings.

[0403] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the hardware environment for training and deploying the model.

[0404] In some embodiments, the effectiveness of the AI ​​model within the scope of use indicated by the first information is guaranteed by optimizing the software environment for training and deploying the model.

[0405] The communication method involved in the embodiments of this disclosure may include at least one of steps S4301 to S4302. For example, step S4301 may be implemented as a standalone embodiment, step S4302 may be implemented as a standalone embodiment, and step S4301+S4302 may be implemented as a standalone embodiment, but is not limited thereto.

[0406] In some embodiments, step S4301 is optional and may be omitted or replaced in different embodiments.

[0407] In some embodiments, step S4302 is optional and may be omitted or replaced in different embodiments.

[0408] According to the solution provided in the embodiments of this disclosure, a new quantization parameter can be introduced to indicate the generalization of the AI ​​model. Optionally, the quantization parameter indicates the scope of use of the AI ​​model, which can be indicated based on the location and / or time of use of the AI ​​model.

[0409] The location used can be a community, geographical coordinates, path, etc., but is not limited to these.

[0410] In some embodiments, quantization parameters may also be introduced to indicate the proportion that the AI ​​model should maintain within the scope of use.

[0411] The quantization parameters can be used for AI model training (such as model training algorithm selection), data collection and / or filtering for model training, and scheduling of computing resources.

[0412] The quantization parameter can be used to identify the AI ​​training service and can be included in the service request.

[0413] In some embodiments, a first communication device sends first information to a second communication device, the first information indicating the scope of use of the AI ​​model, so that the second communication device can ensure the effective coverage of the AI ​​model within the scope of use based on the first information.

[0414] In some embodiments, if the second communication device lacks AI computing power resources or has insufficient AI computing power resources, the effective coverage of the AI ​​model within the scope of use based on the first information provided by the second communication device may be as follows:

[0415] The third communication device sends capability information (including time and location dimensions) that can ensure effective AI coverage to the second communication device. The third communication device ensures the effective coverage of the AI ​​model within the scope of use based on the third communication device's capability information and the capability indication of the third communication device.

[0416] The second communication device sends an AI computing power resource request message (which may include specific AI computing power resource requirements or the first message) to the third communication device to request the third communication device to provide corresponding AI computing power resources to ensure the effective coverage of the AI ​​model within the scope of use.

[0417] In some embodiments, the second communication device has AI computing power resources or the AI ​​computing power resources are sufficient to meet the needs of the first communication device. The second communication device receives the first information sent by the first communication device, and the second communication device ensures the effective coverage of the AI ​​model within the scope of use based on its internal resource scheduling configuration.

[0418] In some embodiments, the assessment of effective coverage may be determined based on an assessment implemented within the second communication device or by additional effectiveness assessment parameters.

[0419] In some embodiments, the second communication device's effective coverage of the AI ​​model within the range based on the first information may involve selecting a model training algorithm based on the first information, or collecting and / or filtering data at specific times and / or locations for model training based on the first information, or scheduling computing resources based on the first information.

[0420] In some embodiments, the effective coverage of the second communication device based on the first information assurance AI model within the stated range can also be achieved based on at least one of the following:

[0421] Improving data quality: The accuracy, completeness, and consistency of data have a direct impact on the generalization ability of models;

[0422] Increase data diversity: Training data should cover all possible input scenarios to avoid overfitting the model to specific situations;

[0423] Increase the amount of data: Sufficient data helps the model learn a wider range of features and improves its generalization ability;

[0424] Optimize data distribution: The distribution of training data should be similar to the distribution of data in real-world applications to ensure that the model can generalize to new data;

[0425] Optimizing model selection: Choosing an appropriate model architecture is crucial for improving generalization. An overly complex model may lead to overfitting, while an overly simple model may fail to capture the complexity of the data.

[0426] Continuous learning: After model deployment, continuously collect new data and update the model to adapt to changes in data distribution;

[0427] Optimize feature extraction: Effective feature selection and feature transformation can improve the generalization ability of the model;

[0428] Ensemble learning: Combining the prediction results of multiple models can improve the generalization ability of the model;

[0429] Using regularization techniques, such as L1 regularization, L2 regularization, and Dropout, can reduce model complexity and improve generalization.

[0430] Optimizing training methods, such as gradient descent and optimizer selection, can affect the training process and final performance of the model.

[0431] Optimize hyperparameter settings: The settings of hyperparameters such as learning rate, batch size, and number of iterations will affect the training effect and generalization ability of the model;

[0432] Optimize hardware and software environment: The hardware and software environment used to train and deploy the model can also affect the model's generalization ability.

[0433] In some embodiments, the first information may also indicate the scope of use validity information of the AI ​​model, which represents the proportion of the AI ​​model that must be effective within the scope of use.

[0434] In some embodiments, the validity information includes the effective proportion of the AI ​​model over time, meaning that the AI ​​model needs to be available within the effective proportion of the time frame during the usage period of the AI ​​model.

[0435] In some embodiments, the validity information includes the effective proportion of the AI ​​model in the location dimension, that is, within the location area where the AI ​​model is used, it is necessary to ensure that the AI ​​model is usable within the location area of ​​the effective proportion.

[0436] In some embodiments, the first information may indicate the QoS requirements of the service, which may be indicated by an identifier, and the second communication device may determine the scope of use of the AI ​​model and / or the validity information of the scope of use of the AI ​​model based on the QoS requirements.

[0437] In some embodiments, the first information may also be directly indicated to the second communication device as a performance requirement parameter.

[0438] In some embodiments, the first communication device may be an AI service consumer or an AI model consumer.

[0439] In some embodiments, the second communication device may be an AI control and management node or an AI model training node.

[0440] In some embodiments, the first information may be associated with AI services, including but not limited to AI model training services or AI inference services.

[0441] In some embodiments, the first information may be sent to the second communication device in a service request.

[0442] In some embodiments, the scope of use of the AI ​​model is determined based on at least one of the following indications:

[0443] Location information used by AI models;

[0444] Information on the usage time of AI models.

[0445] In some embodiments, the location information used by the AI ​​model can be determined based on at least one of the following:

[0446] Community signage information;

[0447] Geographic coordinate information;

[0448] Movement path information.

[0449] Figure 5A is a flowchart illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 5A, the embodiments of the present disclosure relate to an AI model configuration method, which includes:

[0450] Step S5101: Send the first message.

[0451] The optional implementations of step S5101 can be found in the optional implementations of step S3101 in Figure 3, step S4101 in Figure 4A, step S4201 in Figure 4B, step S4301 in Figure 4C, and other related parts in the embodiments involved in Figures 3, 4A, 4B, and 4C, which will not be repeated here.

[0452] In some embodiments, the first communication device sends first information to the second communication device, but is not limited thereto; it may also send first information to other entities.

[0453] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0454] In some embodiments, the first information is also used by the second communication device to ensure the effectiveness of the AI ​​model within its scope of use.

[0455] The communication method involved in the embodiments of this disclosure may include at least step S5101, and step S5101 may be implemented as a standalone embodiment, but is not limited thereto.

[0456] Figure 5B is a flowchart illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 5B, the present disclosure relates to an AI model configuration method, which includes:

[0457] Step S5201: Obtain the first information.

[0458] The optional implementation of step S5201 can be found in the optional implementation of step S3101 in Figure 3, the optional implementation of step S4101 in Figure 4A, and other related parts in the embodiments involved in Figures 3 and 4A, which will not be repeated here.

[0459] In some embodiments, the second communication device receives the first information sent by the first communication device, but is not limited thereto; it may also receive the first information sent by other entities.

[0460] In some embodiments, the second communication device acquires first information as defined by the protocol.

[0461] In some embodiments, the second communication device obtains the first information from the upper layer(s).

[0462] In some embodiments, the second communication device processes the information to obtain the first information.

[0463] In some embodiments, step S5201 is omitted, and the second communication device autonomously implements the function indicated by the first information, or the above function is a default or default setting.

[0464] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0465] Step S5202: Obtain capability information.

[0466] The optional implementation of step S5202 can be found in the optional implementation of step S3102 in Figure 3, the optional implementation of step S4102 in Figure 4A, and other related parts in the embodiments involved in Figures 3 and 4A, which will not be repeated here.

[0467] In some embodiments, the second communication device receives capability information sent by the third communication device, but is not limited thereto; it may also receive capability information sent by other entities.

[0468] In some embodiments, the second communication device acquires capability information specified by the protocol.

[0469] In some embodiments, the second communication device obtains capability information from the upper layer(s).

[0470] In some embodiments, the second communication device processes the information to obtain capability information.

[0471] In some embodiments, step S5202 is omitted, and the second communication device autonomously implements the function indicated by the capability information, or the above function is defaulted or set to default.

[0472] In some embodiments, capability information is used to indicate the computing resources provided by the third communication device, and / or, capability information is used to indicate the scope of use by which the third communication device can provide guarantees for the AI ​​model.

[0473] Step S5203: Obtain computing resources.

[0474] The optional implementation of step S5202 can be found in the optional implementation of step S3102 in Figure 3, the optional implementation of step S4103 in Figure 4A, and other related parts in the embodiments involved in Figures 3 and 4A, which will not be repeated here.

[0475] In some embodiments, the second communication device receives computing resources provided by the third communication device, but is not limited thereto; it may also receive computing resources provided by other entities.

[0476] In some embodiments, the second communication device acquires computing resources as specified in the protocol.

[0477] In some embodiments, the second communication device obtains computing resources from the upper layer(s).

[0478] In some embodiments, the second communication device performs processing to obtain computing resources.

[0479] In some embodiments, step S5203 is omitted, and the second communication device autonomously implements the function indicated by the computing resources, or the above function is default or default.

[0480] In some embodiments, computing resources are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0481] Step S5204: Based on the first information and capability information, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured by the acquired computing resources.

[0482] The optional implementation of step S5202 can be found in the optional implementation of step S3102 in Figure 3, the optional implementation of step S4104 in Figure 4A, and other related parts in the embodiments involved in Figures 3 and 4A, which will not be repeated here.

[0483] The communication method involved in the embodiments of this disclosure may include at least one of steps S5201 to S5204. For example, step S5201 can be implemented as an independent embodiment, step S5204 can be implemented as an independent embodiment, step S5201+S5202 can be implemented as an independent embodiment, step S5201+S5203 can be implemented as an independent embodiment, step S5201+S5204 can be implemented as an independent embodiment, step S5202+S5204 can be implemented as an independent embodiment, step S5203+S5204 can be implemented as an independent embodiment, step S5201+S5202+S5203 can be implemented as an independent embodiment, step S5201+S5202+S5204 can be implemented as an independent embodiment, step S5201+S5203+S5204 can be implemented as an independent embodiment, but not limited thereto.

[0484] In some embodiments, steps S5201 and S5202 can be swapped or executed simultaneously, and steps S5201 and S5203 can be swapped or executed simultaneously.

[0485] In some embodiments, steps S5202, S5203, and S5204 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0486] In some embodiments, steps S5201, S5202, and S5203 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0487] In this embodiment of the disclosure, step S5201 can be combined with step S5101 of FIG5A.

[0488] Figure 5C is a flowchart illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 5C, the embodiments of the present disclosure relate to an AI model configuration method, which includes:

[0489] Step S5301: Obtain the first information.

[0490] The optional implementation of step S5301 can be found in the optional implementation of step S3101 in Figure 3, the optional implementation of step S4201 in Figure 4B, and other related parts in the embodiments involved in Figures 3 and 4B, which will not be repeated here.

[0491] In some embodiments, the second communication device receives the first information sent by the first communication device, but is not limited thereto; it may also receive the first information sent by other entities.

[0492] In some embodiments, the second communication device acquires first information as defined by the protocol.

[0493] In some embodiments, the second communication device obtains the first information from the upper layer(s).

[0494] In some embodiments, the second communication device processes the information to obtain the first information.

[0495] In some embodiments, step S5301 is omitted, and the second communication device autonomously implements the function indicated by the first information, or the above function is default or default.

[0496] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0497] Step S5302: Send resource request information.

[0498] The optional implementation of step S5302 can be found in the optional implementation of step S3101 in Figure 3, the optional implementation of step S4202 in Figure 4B, and other related parts in the embodiments involved in Figures 3 and 4B, which will not be repeated here.

[0499] In some embodiments, the second communication device sends resource request information to the third communication device, but is not limited thereto; it may also send resource request information to other entities.

[0500] In some embodiments, the resource request information is used to request a third communication device to provide computing resources.

[0501] Step S5303: Obtain computing resources.

[0502] The optional implementation of step S5303 can be found in the optional implementation of step S3101 in Figure 3, the optional implementation of step S4203 in Figure 4B, and other related parts in the embodiments involved in Figures 3 and 4B, which will not be repeated here.

[0503] In some embodiments, the second communication device receives computing resources provided by the third communication device, but is not limited thereto; it may also receive computing resources provided by other entities.

[0504] In some embodiments, the second communication device acquires computing resources as specified in the protocol.

[0505] In some embodiments, the second communication device obtains computing resources from the upper layer(s).

[0506] In some embodiments, the second communication device performs processing to obtain computing resources.

[0507] In some embodiments, step S5203 is omitted, and the second communication device autonomously implements the function indicated by the computing resources, or the above function is default or default.

[0508] In some embodiments, computing resources are used to ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0509] Step S5304: By acquiring computing resources, ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0510] The optional implementation of step S5304 can be found in the optional implementation of step S3101 in Figure 3, the optional implementation of step S4204 in Figure 4B, and other related parts in the embodiments involved in Figures 3 and 4B, which will not be repeated here.

[0511] The communication method involved in the embodiments of this disclosure may include at least one of steps S5301 to S5304. For example, step S5301 may be implemented as an independent embodiment, step S5304 may be implemented as an independent embodiment, step S5301+S5302 may be implemented as an independent embodiment, step S5301+S5303 may be implemented as an independent embodiment, step S5301+S5304 may be implemented as an independent embodiment, step S5303+S5304 may be implemented as an independent embodiment, step S5301+S5302+S5303 may be implemented as an independent embodiment, step S5301+S5302+S5304 may be implemented as an independent embodiment, and step S5302+S5303+S5304 may be implemented as an independent embodiment, but is not limited thereto.

[0512] In some embodiments, steps S5302, S5303, and S5304 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0513] In some embodiments, steps S5301, S5302, and S5303 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0514] In this embodiment of the disclosure, step S5301 can be combined with step S5101 of FIG5A.

[0515] Figure 5D is a flowchart illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 5D, the embodiments of the present disclosure relate to an AI model configuration method, which includes:

[0516] Step S5401: Obtain the first information.

[0517] The optional implementation of step S5401 can be found in the optional implementation of step S3101 in Figure 3, the optional implementation of step S4301 in Figure 4C, and other related parts in the embodiments involved in Figures 3 and 4C, which will not be repeated here.

[0518] In some embodiments, the second communication device receives the first information sent by the first communication device, but is not limited thereto; it may also receive the first information sent by other entities.

[0519] In some embodiments, the second communication device acquires first information as defined by the protocol.

[0520] In some embodiments, the second communication device obtains the first information from the upper layer(s).

[0521] In some embodiments, the second communication device processes the information to obtain the first information.

[0522] In some embodiments, step S5401 is omitted, and the second communication device autonomously implements the function indicated by the first information, or the above function is a default or default setting.

[0523] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0524] Step S5402: Based on the first information, the effectiveness of the AI ​​model within the scope of use indicated by the first information is ensured through internal resource scheduling and configuration.

[0525] The optional implementation of step S5402 can be found in the optional implementation of step S3102 in Figure 3, the optional implementation of step S4302 in Figure 4C, and other related parts in the embodiments involved in Figures 3 and 4C, which will not be repeated here.

[0526] The communication method involved in the embodiments of this disclosure may include at least one of steps S5401 to S5402. For example, step S5401 may be implemented as a standalone embodiment, step S5402 may be implemented as a standalone embodiment, and step S5401+S5402 may be implemented as a standalone embodiment, but is not limited thereto.

[0527] In some embodiments, step S5401 is optional and may be omitted or replaced in different embodiments.

[0528] In some embodiments, step S5402 is optional and may be omitted or replaced in different embodiments.

[0529] In this embodiment of the disclosure, step S5401 can be combined with step S5101 of FIG5A.

[0530] Figure 5E is a flowchart illustrating an AI model configuration method according to an embodiment of the present disclosure. As shown in Figure 5E, this disclosure relates to an AI model configuration method, which includes:

[0531] Step S5501: Obtain the first information.

[0532] The optional implementations of step S5501 can be found in the optional implementations of step S3101 in Figure 3, step S4101 in Figure 4A, step S4201 in Figure 4B, step S4301 in Figure 4C, step S5201 in Figure 5B, step S5301 in Figure 5C, step S5401 in Figure 5D, and other related parts in the embodiments involved in Figures 3, 4A, 4B, 4C, 5B, 5C, and 5D, which will not be repeated here.

[0533] In some embodiments, the second communication device receives the first information sent by the first communication device, but is not limited thereto; it may also receive the first information sent by other entities.

[0534] In some embodiments, the second communication device acquires first information as defined by the protocol.

[0535] In some embodiments, the second communication device obtains the first information from the upper layer(s).

[0536] In some embodiments, the second communication device processes the information to obtain the first information.

[0537] In some embodiments, step S5501 is omitted, and the second communication device autonomously implements the function indicated by the first information, or the above function is a default or default setting.

[0538] In some embodiments, the first information is used to indicate the scope of use of the AI ​​model.

[0539] Step S5502: Based on the first information, ensure the effectiveness of the AI ​​model within the scope of use indicated by the first information.

[0540] The optional implementations of step S5502 can be found in the optional implementations of step S3102 in Figure 3, the optional implementations of steps S4102, S4103, and S4104 in Figure 4A, the optional implementations of steps S4202, S4203, and S4204 in Figure 4B, the optional implementation of step S4302 in Figure 4C, the optional implementations of steps S5202, S5203, and S5204 in Figure 5B, the optional implementations of steps S5302, S5303, and S5304 in Figure 5C, the optional implementation of step S5402 in Figure 5D, and other related parts in the embodiments involved in Figures 3, 4A, 4B, 4C, 5B, 5C, and 5D, which will not be repeated here.

[0541] The communication method involved in the embodiments of this disclosure may include at least one of steps S5401 to S5402. For example, step S5401 may be implemented as a standalone embodiment, step S5402 may be implemented as a standalone embodiment, and step S5401+S5402 may be implemented as a standalone embodiment, but is not limited thereto.

[0542] In some embodiments, step S5401 is optional and may be omitted or replaced in different embodiments.

[0543] In some embodiments, step S5402 is optional and may be omitted or replaced in different embodiments.

[0544] In this embodiment of the disclosure, step S5501 can be combined with step S5101 of FIG5A.

[0545] In the embodiments disclosed herein, some or all of the steps and their optional implementations may be arbitrarily combined with some or all of the steps in other embodiments, or may be arbitrarily combined with the optional implementations in other embodiments.

[0546] This disclosure also provides embodiments of an apparatus for implementing any of the above methods. For example, an apparatus is provided that includes units or modules for implementing the steps performed by the first communication device in any of the above methods. Furthermore, another apparatus is provided that includes units or modules for implementing the steps performed by the second communication device in any of the above methods.

[0547] It should be understood that the division of units or modules in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, the units or modules in the device can be implemented by a processor calling software: for example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of the units or modules in the above device. The processor can be, for example, a general-purpose processor, such as a Central Processing Unit (CPU) or a microprocessor, and the memory can be internal or external to the device. Alternatively, the units or modules in the device can be implemented in the form of hardware circuits. The functionality of some or all of the units or modules can be achieved through the design of these hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC). The functionality of some or all of the units or modules is achieved through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a programmable logic device (PLD). Taking a field-programmable gate array (FPGA) as an example, it can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files, thereby achieving the functionality of some or all of the units or modules. All units or modules of the above device can be implemented entirely through processor-called software, entirely through hardware circuits, or partially through processor-called software with the remaining parts implemented through hardware circuits.

[0548] In this embodiment, the processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a Central Processing Unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. The logical relationships of the aforementioned hardware circuits are fixed or reconfigurable. For example, the processor is a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units or modules. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a Neural Network Processing Unit (NPU), a Tensor Processing Unit (TPU), or a Deep Learning Processing Unit (DPU).

[0549] Figure 6A is a schematic diagram of the structure of a first communication device according to an embodiment of this disclosure. As shown in Figure 6A, the first communication device 6100 may include at least a transceiver module 6101. In some embodiments, the transceiver module 6101 is configured to send first information to a second communication device, the first information being used to indicate the scope of use of an AI model, and the first information being used by the second communication device to ensure the effectiveness of the AI ​​model within the scope of use. Optionally, the transceiver module 6101 is used to perform at least one of the communication steps (e.g., step S3101, but not limited thereto) performed by the first communication device in any of the above methods, which will not be elaborated here. In some embodiments, the first communication device 6100 may further include a processing module. Optionally, the processing module is used to perform at least one of the other steps performed by the first communication device in any of the above methods, which will not be elaborated here.

[0550] Figure 6B is a schematic diagram of the structure of the second communication device proposed in an embodiment of this disclosure. As shown in Figure 6B, the second communication device 6200 may include at least one of a transceiver module 6201 and a processing module 6202. In some embodiments, the transceiver module 6201 is configured to receive first information sent by the first communication device, the first information being used to indicate the scope of use of the AI ​​model; the processing module 6202 is configured to ensure the effectiveness of the AI ​​model within the scope of use based on the first information. Optionally, the transceiver module 6201 is used to perform at least one of the communication steps (e.g., step S3101, but not limited thereto) performed by the second communication device in any of the above methods, which will not be described in detail here; the processing module 6202 is used to perform at least one of the other steps (e.g., step S3102, but not limited thereto) performed by the second communication device in any of the above methods, which will not be described in detail here.

[0551] In some embodiments, the transceiver module may include a transmitting module and / or a receiving module, which may be separate or integrated. Optionally, the transceiver module may be interchangeable with a transceiver.

[0552] In some embodiments, the processing module may be a single module or may include multiple sub-modules. Optionally, the multiple sub-modules may each perform all or part of the steps required by the processing module. Optionally, the processing module may be interchangeable with a processor.

[0553] Figure 7A is a schematic diagram of the structure of the communication device 7100 proposed in an embodiment of this disclosure. The communication device 7100 can be a first communication device, a second communication device, a chip, chip system, or processor that supports the first communication device in implementing any of the above methods, or a chip, chip system, or processor that supports the second communication device in implementing any of the above methods. The communication device 7100 can be used to implement the methods described in the above method embodiments; for details, please refer to the descriptions in the above method embodiments.

[0554] As shown in Figure 7A, the communication device 7100 includes one or more processors 7101. The processor 7101 can be a general-purpose processor or a dedicated processor, such as a baseband processor or a central processing unit (CPU). The baseband processor can be used to process communication protocols and communication data, while the CPU can be used to control communication devices (e.g., base stations, baseband chips, terminal devices, terminal device chips, DUs or CUs, etc.), execute programs, and process program data. The communication device 7100 is used to execute any of the above methods.

[0555] In some embodiments, the communication device 7100 further includes one or more memories 7102 for storing instructions. Optionally, all or part of the memories 7102 may also be located outside the communication device 7100.

[0556] In some embodiments, the communication device 7100 further includes one or more transceivers 7103. When the communication device 7100 includes one or more transceivers 7103, the transceivers 7103 perform at least one of the communication steps such as sending and / or receiving in the above method (e.g., step S3101, but not limited thereto), and the processor 7101 performs at least one of the other steps (e.g., step S3102, but not limited thereto).

[0557] In some embodiments, a transceiver may include a receiver and / or a transmitter, which may be separate or integrated. Optionally, the terms transceiver, transceiver unit, transceiver, transceiver circuit, etc., may be used interchangeably; the terms transmitter, transmitting unit, transmitter, transmitting circuit, etc., may be used interchangeably; and the terms receiver, receiving unit, receiver, receiving circuit, etc., may be used interchangeably.

[0558] In some embodiments, the communication device 7100 may include one or more interface circuits 7104. Optionally, the interface circuit 7104 is connected to the memory 7102, and the interface circuit 7104 can be used to receive signals from the memory 7102 or other devices, and can be used to send signals to the memory 7102 or other devices. For example, the interface circuit 7104 can read instructions stored in the memory 7102 and send the instructions to the processor 7101.

[0559] The communication device 7100 described in the above embodiments may be a first communication device or a second communication device, but the scope of the communication device 7100 described in this disclosure is not limited thereto, and the structure of the communication device 7100 may not be limited by FIG. 7A. The communication device may be a standalone device or may be part of a larger device. For example, the communication device may be: (1) a standalone integrated circuit IC, or chip, or chip system or subsystem; (2) a collection of one or more ICs, optionally, the IC collection may also include storage components for storing data and programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, terminal device, smart terminal device, cellular phone, wireless device, handheld device, mobile unit, vehicle device, network device, cloud device, artificial intelligence device, etc.; (6) others, etc.

[0560] Figure 7B is a schematic diagram of the structure of the chip 7200 according to an embodiment of this disclosure. For cases where the communication device 7100 can be a chip or a chip system, the schematic diagram of the chip 7200 shown in Figure 7B can be referred to, but is not limited thereto.

[0561] Chip 7200 includes one or more processors 7201, which are used to perform any of the above methods.

[0562] In some embodiments, chip 7200 further includes one or more interface circuits 7202. Optionally, the interface circuit 7202 is connected to memory 7203, and the interface circuit 7202 can be used to receive signals from memory 7203 or other devices, and the interface circuit 7202 can be used to send signals to memory 7203 or other devices. For example, the interface circuit 7202 can read instructions stored in memory 7203 and send the instructions to processor 7201.

[0563] In some embodiments, the interface circuit 7202 performs at least one of the communication steps such as sending and / or receiving in the above method (e.g., step S3101, but not limited thereto), and the processor 7201 performs at least one of the other steps (e.g., step S3102, but not limited thereto).

[0564] In some embodiments, the terms interface circuit, interface, transceiver pin, transceiver, etc., can be used interchangeably.

[0565] In some embodiments, chip 7200 further includes one or more memories 7203 for storing instructions. Optionally, all or part of the memories 7203 may be located outside of chip 7200.

[0566] This disclosure also proposes a storage medium storing instructions that, when executed on the communication device 7100, cause the communication device 7100 to perform any of the above methods. Optionally, the storage medium is an electronic storage medium. Optionally, the storage medium is a computer-readable storage medium, but not limited thereto; it may also be a storage medium readable by other devices. Optionally, the storage medium may be a non-transitory storage medium, but not limited thereto; it may also be a temporary storage medium.

[0567] This disclosure also proposes a program product, including a computer program, which, when executed by the communication device 7100, causes the communication device 7100 to perform any of the above methods. Alternatively, this disclosure also proposes a program product that, when executed by the communication device 7100 or run on the communication device 7100, causes the communication device 7100 to perform any of the above methods. Optionally, the above program product is a computer program product.

[0568] This disclosure also proposes a computer program that, when run on a computer, causes the computer to perform any of the above methods.

[0569] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0570] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for configuring an artificial intelligence (AI) model, characterized in that, Applied to a first communication device, the method comprises: sending first information to a second communication device, the first information being used to indicate a usage range of an AI model, and the first information also being used to ensure the validity of the AI model within the usage range.

2. The method of claim 1, wherein, The usage range of the AI model comprises at least one of: a usage area of the AI model; a usage time of the AI model.

3. The method of claim 2, wherein, The usage area of the AI model is determined based on at least one of: cell identification information; geographic coordinate information; mobile path information.

4. The method according to any one of claims 1 to 3, characterized in that, The first information is also used to indicate validity information of the usage range, and the validity information is used to indicate the proportion of the usage range that the AI model needs to ensure effective within the usage range indicated by the first information.

5. The method of claim 4, wherein, The validity information is used to indicate at least one of: the proportion of the usage time that the AI model needs to ensure effective within the usage time indicated by the first information; the proportion of the usage area that the AI model needs to ensure effective within the usage area indicated by the first information.

6. The method according to any one of claims 1 to 5, characterized in that, The first information is indicated by at least one of: performance requirement parameters; quality of service (QoS) requirement information, which is used to determine the first information.

7. The method according to any one of claims 1 to 6, characterized in that, The first information is associated with an AI service, and the AI service comprises at least one of: AI model training service; AI model inference service; AI model data service.

8. The method according to any one of claims 1 to 7, characterized in that, The first information is used by the second communication device to ensure the validity of the AI model within the usage range indicated by the first information based on the first information and capability information of a third communication device, and through the computing resource provided by the third communication device, wherein the capability information is used to indicate the computing resource provided by the third communication device, and / or the capability information is used to indicate the usage range that the third communication device can guarantee for the AI model; or The first information is used by the second communication device to send resource request information to a third communication device, and the resource request information is used to request the third communication device to provide computing resource, and the computing resource is used to ensure the validity of the AI model within the usage range indicated by the first information.

9. The method according to any one of claims 1 to 8, characterized in that, The validity of the AI model within the usage range by the second communication device comprises at least one of: The first information is used by the second communication device to select a model training algorithm according to the first information; The first information is used by the second communication device to collect first data for model training or model inference according to the first information, and the first data is data of a specific time and / or data of a specific location; The first information is used by the second communication device to filter first data for model training or model inference according to the first information, and the first data is data of a specific time and / or data of a specific location; The first information is used by the second communication device to schedule computing resource according to the first information, and the computing resource is used to ensure the validity of the AI model within the usage range.

10. The method according to any one of claims 1 to 9, characterized in that, The first communication device is a communication device using the AI model, and / or the first communication device is a communication device using an AI service associated with the first information.

11. The method according to any one of claims 1 to 10, characterized in that, The second communication device is configured to provide at least one of the following functions: a control management function of the AI model; a training function of the AI model; a data function of the AI model. 12.A configuration method of an AI model, comprising: The method applied to the second communication device comprises: receiving first information sent by a first communication device, the first information being used to indicate a usage range of an AI model; based on the first information, ensuring the effectiveness of the AI model within the usage range.

13. The method of claim 12, wherein, The usage range of the AI model comprises at least one of the following: a usage area of the AI model; a usage time of the AI model.

14. The method of claim 13, wherein, The usage area of the AI model is determined based on at least one of the following: cell identification information; geographical coordinate information; mobile path information.

15. The method according to any one of claims 12 to 14, characterized in that, The first information is also used to indicate validity information of the usage range, and the validity information is used to indicate a proportion of a usage range that the AI model needs to ensure effective within the usage range indicated by the first information.

16. The method of claim 15, wherein, The validity information is used to indicate at least one of the following: a proportion of a usage time that the AI model needs to ensure effective within the usage time indicated by the first information; a proportion of a usage area that the AI model needs to ensure effective within the usage area indicated by the first information.

17. The method according to any one of claims 12 to 16, characterized in that, The first information is indicated by at least one of the following: performance requirement parameters; quality of service (QoS) requirement information, which is used to determine the first information.

18. The method according to any one of claims 12 to 17, characterized in that, The first information is associated with an AI service, and the AI service comprises at least one of the following: AI model training service; AI model inference service; AI model data service.

19. The method according to any one of claims 12 to 18, characterized in that, Based on the first information, the effectiveness of the AI model within the usage range is ensured by: receiving capability information sent by a third communication device, the capability information being used to indicate computing resource provided by the third communication device, and / or the capability information being used to indicate a usage range that the third communication device can guarantee for the AI model; based on the capability information and the first information, the effectiveness of the AI model within the usage range indicated by the first information is ensured by the computing resource provided by the third communication device.

20. The method of any one of claims 12-18, wherein, Based on the first information, the effectiveness of the AI model within the usage range is ensured by: based on the first information, sending resource request information to a third communication device, the resource request information being used to request the third communication device to provide computing resource, and the computing resource being used to ensure the effectiveness of the AI model within the usage range indicated by the first information.

21. The method according to any one of claims 12 to 20, characterized in that, The effectiveness of the AI model within the usage range is determined by the second communication device according to internal implementation, or the effectiveness of the AI model within the usage range is determined by the second communication device according to effectiveness evaluation parameters obtained from the outside.

22. The method of any one of claims 12-21, wherein, The effectiveness of the AI model within the usage range comprises at least one of the following: According to the first information, a model training algorithm is selected; According to the first information, first data for model training or model inference is collected, the first data being data of a specific time and / or data of a specific location; According to the first information, first data for model training or model inference is screened, the first data being data of a specific time and / or data of a specific location; According to the first information, an algorithm resource is scheduled, the algorithm resource being used to ensure effectiveness of the AI model within the use range.

23. The method of any one of claims 12-22, wherein, The first communication device is a communication device using the AI model, and / or the first communication device is a communication device using an AI service associated with the first information.

24. The method of any one of claims 12-23, wherein, The second communication device is used to provide at least one of the following functions: A control management function of the AI model; A training function of the AI model; A data function of the AI model.

25. A first communication device, characterized by Comprising: A transceiver module configured to send first information to a second communication device, the first information being used to indicate a use range of an AI model, and the first information also being used by the second communication device to ensure effectiveness of the AI model within the use range.

26. A second communication device, characterized by Comprising: A transceiver module configured to receive first information sent by a first communication device, the first information being used to indicate a use range of an AI model; A processing module configured to ensure effectiveness of the AI model within the use range based on the first information.

27. A first communication device, the first communication device comprising: Comprising: One or more processors; The first communication device is used to perform the AI model configuration method of any one of claims 1-11.

28. A second communication device, characterized by Comprising: One or more processors; The second communication device is used to perform the AI model configuration method of any one of claims 12-24.

29. A communication system, characterized by Comprising a first communication device and a second communication device, wherein the first communication device is configured to implement the AI model configuration method of any one of claims 1-11, and the second communication device is configured to implement the AI model configuration method of any one of claims 12-24.

30. A storage medium, the storage medium storing instructions, wherein, When the instructions run on a communication device, the communication device is caused to perform the AI model configuration method of any one of claims 1-11 or 12-24.

31. A program product, characterized by The program product comprises a computer program, when the computer program is executed by a communication device, the communication device is caused to perform the AI model configuration method of any one of claims 1-11 or 12-24.

Images