Communication method and related device

Abstract

The embodiment of the invention provides a communication method and a related device, which are used for reducing the signaling overhead of a first device for sending local model parameters of a first model. A first device receives first information from a second device, wherein the first information is used for respectively indicating whether the first device sends each local model parameter of a first model of the first device; the first device determines a part of local model parameters of the first model to be sent according to the first information, wherein the part of local model parameters are obtained by training the first model; and the first device sends the partial local model parameters to the second device.

Description

A communication method and related device Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a communication method and related devices. Background Art

[0002] Starting with release 16 (R16), the fifth-generation (5G) mobile communication system (5G) network has been researching the use of network data analysis functionality (NWDAF) elements to support artificial intelligence (AI) capabilities within 5G networks. NWDAF elements are primarily used for data collection and analysis at the application layer, providing external services and interface calls. In R18, research projects are underway to expand the functionality of NWDAF elements, enabling support for external AI services and model transmission within the network.

[0003] The integration of AI and networks will be a key research direction in the future. Model-related parameters require extensive network transmission. As models grow in size, so do the number of model-related parameters. Consequently, transmitting these parameters over wireless networks incurs significant signaling overhead. Therefore, reducing the signaling overhead associated with transmitting model-related parameters between devices is a worthy challenge.

[0004] Summary of the Invention

[0005] The present application provides a communication method and related devices for reducing the signaling overhead of a first device sending local model parameters of a first model.

[0006] A first aspect of the present application provides a communication method, comprising:

[0007] The first device receives first information from the second device, where the first information is used to indicate whether the first device should send the local model parameters of the first model of the first device; the first device determines part of the local model parameters of the first model to be sent based on the first information, where the part of the local model parameters is obtained by training the first model; the first device sends the part of the local model parameters to the second device.

[0008] As can be seen from the above technical solution, the first device can determine some local model parameters of the first model based on the first information. The first device then sends these local model parameters to the second device. The terminal device does not need to send all the local model parameters of the first model. This reduces the signaling overhead of the first device reporting the local model parameters of the first model. Furthermore, the first device can calculate only some of the local model parameters of the first model, eliminating the need to calculate the local model parameters of the first model that do not need to be sent. This reduces the computational effort of the first device and reduces energy consumption.

[0009] A second aspect of the present application provides a communication method, including:

[0010] The second device sends first information to the first device, and the first information is used to indicate whether the first device sends the local model parameters of the first model of the first device; the second device receives part of the local model parameters of the first model from the first device, and the part of the local model parameters is obtained by training the first model.

[0011] As can be seen from the above technical solution, a first device can send first information to a second device, thereby instructing the first device to transmit partial local model parameters of a first model of the first device. The first device can transmit partial local model parameters of the first model to the second device. The terminal device does not need to transmit all local model parameters of the first model. This reduces the signaling overhead of the first device reporting the local model parameters of the first model. Furthermore, the first device can calculate only the partial local model parameters of the first model, eliminating the need to calculate the local model parameters of the first model that do not need to be transmitted. This reduces the computational effort of the first device and reduces energy consumption.

[0012] Based on the first aspect or the second aspect, in one possible implementation, the local model parameters include local weight parameters of the first model. In this implementation, a specific form of the local model parameters is shown. The technical solution of the present application enables transmission of the local weight parameters of the first model between the first and second devices, thereby reducing the overhead incurred by transmitting the local weight parameters between the first and second devices.

[0013] Based on the first aspect or the second aspect, in a possible implementation manner, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0014] Based on the first aspect or the second aspect, in a possible implementation method, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

[0015] In this implementation, the first information includes N first indication information, and the N first indication information correspond one-to-one to N local model parameters, so that each first indication information is used to instruct the first device whether to send the local model parameter corresponding to the first indication information.

[0016] Based on the first aspect or the second aspect, in a possible implementation method, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters.

[0017] In this implementation, the first information includes P pieces of second indication information, each of which corresponds one-to-one to the local model parameters of neurons in P layers. Thus, each piece of second indication information is used to instruct the first device whether to transmit the local model parameters of the neurons in the layer corresponding to the second indication information. Furthermore, in this implementation, the second device instructs the first device on a layer-by-layer basis whether to transmit the local model parameters of neurons in each layer, which helps reduce the overhead incurred by the second device in transmitting the first information.

[0018] Based on the first aspect, in a possible implementation manner, the method further includes: the first device receiving N global model parameters of the first model or the global model parameters of P layers of neurons of the first model from the second device.

[0019] In this implementation, the first device may receive N global model parameters of the first model or global model parameters of P layers of neurons, thereby facilitating the first device to update the first model in combination with the N global model parameters of the first model or global model parameters of P layers of neurons.

[0020] Based on the second aspect, in a possible implementation manner, the method further includes: the second device sending N global model parameters of the first model or global model parameters of P layers of neurons in the first model to the first device.

[0021] In this implementation, the second device may send the N global model parameters of the first model or the global model parameters of the P-layer neurons of the first model to the first device, thereby facilitating the first device to update the first model in combination with the N global model parameters of the first model or the global model parameters of the P-layer neurons.

[0022] Based on the first aspect or the second aspect, in a possible implementation method, N global model parameters correspond one-to-one to N local model parameters; N first indication information and N global model parameters are carried in the same signaling or different signaling; when N first indication information and N global model parameters are carried in the same signaling, the N global model parameters and the N first indication information are arranged at intervals, and the first indication information corresponding to each global model parameter is arranged adjacent to the first indication information of the global model parameter, or, the N global model parameters are arranged before the N first indication information.

[0023] In this implementation, the N global model parameters and the N local model parameters can be carried in the same signaling or in different signaling. For the case where the N global model parameters and the N local model parameters are carried in the same signaling, two formats of the N global model parameters and the N local model parameters in the signaling are shown.

[0024] Based on the first aspect or the second aspect, in a possible implementation method, the global model parameters of the P layer neurons correspond one-to-one to the local model parameters of the P layer neurons; the P second indication information and the global model parameters of the P layer neurons are carried in the same signaling or different signalings. When the P second indication information and the global model parameters of the P layer neurons are carried in the same signaling, the global model parameters of the P layer neurons and the P second indication information are arranged at intervals, and the second indication information corresponding to the global model parameters of each layer of neurons is arranged adjacent to the global model parameters of each layer of neurons, or the global model parameters of the P layer neurons are arranged before the P second indication information.

[0025] In this implementation, the P second indication information and the global model parameters of the P layer neurons can be carried in the same signaling or in different signaling. For the case where the P second indication information and the global model parameters of the P layer neurons are carried in the same signaling, two formats of the P second indication information and the global model parameters of the P layer neurons in the signaling are shown.

[0026] Based on the first aspect or the second aspect, in a possible implementation, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device does not send the local model parameters of at least one neuron of the first target layer; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device sends the local model parameters of at least one neuron of the second target layer.

[0027] In this implementation, the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons. The first identification bit is used to uniformly instruct the first device not to send the local model parameters of the neurons of the at least one first target layer. This reduces the indication overhead of the second device. For scenarios with fewer first target layers, the second device can send the first information through this implementation, which is beneficial to further reduce the indication overhead. Alternatively, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons. The second identification bit is used to uniformly instruct the first device to send the local model parameters of the neurons of the at least one second target layer. This reduces the indication overhead of the second device. For scenarios with fewer second target layers, the second device can send the first information through this implementation, which is beneficial to further reduce the indication overhead.

[0028] A third aspect of the present application provides a communication method, including:

[0029] The first device determines part of the local model parameters of the first model of the first device to be sent, and the part of the local model parameters is obtained by training the first model; the first device sends the part of the local model parameters and first information to the second device, and the first information is used to instruct the first device to send the part of the local model parameters.

[0030] As can be seen from the above technical solution, the first device can determine the partial local model parameters of the first model to be transmitted. The first device then transmits the partial local model parameters of the first model and the first information to the second device. The first information is used to instruct the first device to transmit the partial local model parameters of the first model. Therefore, the first device can transmit only the partial local model parameters of the first model, without having to transmit all the local model parameters of the first model. This reduces the signaling overhead of the first device transmitting the local model parameters of the first model. Furthermore, the first device can calculate only the partial local model parameters of the first model, without having to calculate the local model parameters of the first model that do not need to be transmitted. This reduces the computational effort of the first device and reduces energy consumption losses of the first device.

[0031] A fourth aspect of the present application provides a communication method, including:

[0032] The second device receives partial local model parameters and first information of the first model from the first device, where the first information is used to instruct the first device to send the partial local model parameters, which are obtained by training the first model; the second device determines the partial local model parameters based on the first information.

[0033] As can be seen from the above technical solution, the second device receives partial local model parameters of the first model and the first information from the first device. It can be seen that the first device can only send partial local model parameters of the first model, and does not need to send all local model parameters of the first model. This reduces the signaling overhead of the first device sending the local model parameters of the first model. Furthermore, the first device can only calculate the partial local model parameters of the first model, and does not need to calculate the local model parameters of the first model that do not need to be sent. This reduces the computational effort of the first device and reduces energy consumption losses of the first device.

[0034] Based on the third or fourth aspect, in one possible implementation, the portion of local model parameters includes local weight parameters of the first model. In this implementation, a specific form of the local model parameters is shown. The technical solution of the present application enables transmission of the local weight parameters of the first model between the first and second devices, thereby reducing the overhead incurred by transmitting the local weight parameters between the first and second devices.

[0035] Based on the third aspect or the fourth aspect, in a possible implementation manner, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0036] Based on the third aspect or the fourth aspect, in a possible implementation method, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

[0037] In this implementation, the first information includes N first indication information, and the N first indication information correspond one-to-one to N local model parameters, so that each first indication information is used to instruct the first device whether to send the local model parameter corresponding to the first indication information.

[0038] Based on the third aspect or the fourth aspect, in a possible implementation method, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters of the neurons in that layer.

[0039] In this implementation, the first information includes P pieces of second indication information, each of which corresponds one-to-one to the local model parameters of neurons in P layers. Thus, each piece of second indication information is used to instruct the first device whether to transmit the local model parameters of the neurons in the layer corresponding to the second indication information. Furthermore, in this implementation, the second device instructs the first device on a layer-by-layer basis whether to transmit the local model parameters of neurons in each layer, which helps reduce the overhead incurred by the second device in transmitting the first information.

[0040] Based on the third aspect or the fourth aspect, in a possible implementation, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device does not send the local model parameters of at least one neuron of the first target layer; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device sends the local model parameters of at least one neuron of the second target layer.

[0041] In this implementation, the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons. The first identification bit is used to uniformly instruct the first device not to send the local model parameters of the neurons of the at least one first target layer. This reduces the indication overhead of the first device. For scenarios with fewer first target layers, the first device can send the first information through this implementation, which is beneficial to further reduce the indication overhead. Alternatively, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons. The second identification bit is used to uniformly instruct the first device to send the local model parameters of the neurons of the at least one second target layer. This reduces the indication overhead of the first device. For scenarios with fewer second target layers, the first device can send the first information through this implementation, which is beneficial to further reduce the indication overhead.

[0042] Based on the third aspect, in a possible implementation method, the first device determines some local model parameters of the first model of the first device to be sent, including: the first device determines the some local model parameters based on the local model parameters obtained by the first device for the Rth round of training of the first model, the communication link status of the first device, and at least one of the computing power of the first device, and the some local model parameters are obtained by the first device for the R+1th round of training of the first model, where R is an integer greater than or equal to 1.

[0043] This implementation illustrates a possible method for a first device to determine a portion of the local model parameters of a first model to be transmitted. This facilitates the first device to reasonably determine the portion of the local model parameters to be transmitted, enabling the reporting of important local model parameters to the second device as much as possible. This reduces the overhead of the first device reporting local model parameters without affecting the accuracy of the global model parameters determined by the second device.

[0044] A fifth aspect of the present application provides a communication method, including:

[0045] The first device receives part of the first global model parameters of the first model of the first device from the second device; the first device receives first information from the second device, and the first information is used to instruct the second device to send part of the first global model parameters; the first device updates the first model according to the first information and part of the first global model parameters to obtain an updated first model.

[0046] In the above technical solution, the first device can receive a portion of the first global model parameters of the first model and the first information. The first device then updates the first model based on the first information and the portion of the first global model parameters to obtain an updated first model. Therefore, the second device can send only a portion of the first global model parameters of the first model, without having to send all of the first global model parameters to the first device. This reduces the overhead of the second device sending the first global model parameters.

[0047] A sixth aspect of the present application provides a communication method, including:

[0048] The second device sends part of the first global model parameters of the first model of the first device to the first device; the second device sends first information to the first device, and the first information is used to instruct the second device to send part of the first global model parameters.

[0049] In the above technical solution, the second device transmits a portion of the first global model parameters of the first model of the first device and the first information to the first device. This facilitates the first device updating the first model based on the first information and the portion of the first global model parameters to obtain an updated first model. The second device can transmit only a portion of the first global model parameters of the first model, without having to transmit all of the first global model parameters of the first model to the first device. This reduces the overhead of the second device transmitting the first global model parameters.

[0050] Based on the fifth or sixth aspect, in one possible implementation, the portion of first global model parameters includes global weight parameters of the first model. In this implementation, a specific form of the first global model parameters is shown, and the technical solution of the present application is used to transmit the global weight parameters of the first model between the first device and the second device, thereby reducing the overhead generated by transmitting the global weight parameters between the first device and the second device.

[0051] Based on the fifth aspect or the sixth aspect, in a possible implementation, the global weight parameter includes the global weight or global weight gradient of the first model.

[0052] Based on the fifth aspect or the sixth aspect, in a possible implementation method, all first global model parameters of the first model include N first global model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N first global model parameters. The first indication information corresponding to each first global model parameter in the N first global model parameters is used to indicate whether the second device sends the first global model parameter.

[0053] In this implementation, the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N first global model parameters. Thus, the first indication information corresponding to each first global model parameter in the N first global model parameters is used to instruct the second device whether to send the first global model parameter.

[0054] Based on the fifth aspect or the sixth aspect, in a possible implementation method, all the first global model parameters of the first model include the first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the first global model parameters of the P layers of neurons. The second indication information corresponding to the first global model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the second device sends the first global model parameters of each layer of neurons.

[0055] In this implementation, the first information includes P pieces of second indication information, each of which corresponds one-to-one to the first global model parameters of neurons in P layers. Thus, each piece of second indication information is used to instruct the first device whether to transmit the first global model parameters of the neurons in the layer corresponding to the second indication information. Furthermore, in this implementation, the second device instructs the second device on a layer-by-layer basis whether to transmit the first global model parameters of neurons in each layer, which helps reduce the overhead incurred by the second device in transmitting the first information.

[0056] Based on the fifth or sixth aspect, in one possible implementation, all first global model parameters of the first model include first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1;

[0057] The first information includes a first identification bit and a layer sequence number of at least one first target layer in the P layer neurons, the first identification bit being used to instruct the second device not to send the first global model parameter of the at least one first target layer neuron; or

[0058] The first information includes a second identification bit and a layer sequence number of at least one second target layer in the P layer neurons, and the second identification bit is used to instruct the second device to send a first global model parameter of at least one second target layer neuron.

[0059] In this implementation, the first information includes a first identification bit and the layer sequence number of at least one first target layer in the P-layer neurons. The first identification bit is used to uniformly instruct the second device not to send the first global model parameters of the neurons of the at least one first target layer. This reduces the indication overhead of the second device. For scenarios with fewer first target layers, the second device can send the first information through this implementation, which is beneficial to further reduce the indication overhead. Alternatively, the first information includes a second identification bit and the layer sequence number of at least one second target layer in the P-layer neurons. The second identification bit is used to uniformly instruct the first device to send the first global model parameters of the neurons of the at least one second target layer. This reduces the indication overhead of the second device. For scenarios with fewer second target layers, the second device can send the first information through this implementation, which is beneficial to further reduce the indication overhead.

[0060] Based on the fifth or sixth aspect, in one possible implementation, all first global model parameters of the first model include N first global model parameters obtained by the second device in the M+1 round by fusing the local model parameters of multiple devices, where N is an integer greater than or equal to 2; the N first global model parameters correspond one-to-one with the N second global model parameters, where the N second global model parameters are obtained by the second device in the M round by fusing the local model parameters of multiple devices, where M is an integer greater than or equal to 1; and for some first global model parameters, the ratio of the change between each first global model parameter and the second global model parameter corresponding to the first global model parameter to the second global model parameter is greater than the first ratio. In this implementation, the second device can send first global model parameters with larger changes to the first device, and can discard first global model parameters with smaller changes. This does not affect the accuracy of the first device in updating the first model and can also reduce the reporting overhead of the model parameters.

[0061] A seventh aspect of the present application provides a first device, including:

[0062] a transceiver module, configured to receive first information from a second device, the first information being used to instruct the first device whether to send local model parameters of the first model of the first device;

[0063] a processing module, configured to determine, based on the first information, some local model parameters of the first model to be sent, where the some local model parameters are obtained by training the first model;

[0064] The transceiver module is further configured to send the portion of local model parameters to the second device.

[0065] An eighth aspect of the present application provides a second device, including:

[0066] The transceiver module is used to send first information to the first device, where the first information is used to indicate whether the first device sends the local model parameters of the first model of the first device; and receive some local model parameters of the first model from the first device, where the some local model parameters are obtained by training the first model.

[0067] Based on the seventh aspect or the eighth aspect, in a possible implementation manner, the local model parameters include local weight parameters of the first model.

[0068] Based on the seventh aspect or the eighth aspect, in a possible implementation manner, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0069] Based on the seventh aspect or the eighth aspect, in a possible implementation method, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

[0070] Based on the seventh aspect or the eighth aspect, in a possible implementation method, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters.

[0071] Based on the seventh aspect, in a possible implementation, the transceiver module is further used to: receive N global model parameters of the first model or global model parameters of P layers of neurons of the first model from the second device.

[0072] Based on the eighth aspect, in a possible implementation, the transceiver module is further used to: send N global model parameters of the first model or global model parameters of P layers of neurons of the first model to the first device.

[0073] Based on the seventh aspect or the eighth aspect, in a possible implementation method, N global model parameters correspond one-to-one to N local model parameters; N first indication information and N global model parameters are carried in the same signaling or different signaling. When the N first indication information and N global model parameters are carried in the same signaling, the N global model parameters and the N first indication information are arranged at intervals, and the first indication information corresponding to each global model parameter is arranged adjacent to the global model parameter, or the N global model parameters are arranged before the N first indication information.

[0074] Based on the seventh aspect or the eighth aspect, in a possible implementation method, the global model parameters of the P layer neurons correspond one-to-one to the local model parameters of the P layer neurons; the P second indication information and the global model parameters of the P layer neurons are carried in the same signaling or different signalings. When the P second indication information and the global model parameters of the P layer neurons are carried in the same signaling, the global model parameters of the P layer neurons and the P second indication information are arranged at intervals, and the second indication information corresponding to the global model parameters of each layer of neurons is arranged adjacent to the global model parameters of each layer of neurons, or the global model parameters of the P layer neurons are arranged before the P second indication information, and the interval between the global model parameters of each layer of neurons and the second indication information corresponding to the global model parameters of each layer of neurons is equal; or, the P second indication information and the global model parameters of the P layer neurons are carried in different signalings.

[0075] Based on the seventh aspect or the eighth aspect, in a possible implementation, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device does not send the local model parameters of at least one neuron of the first target layer; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device sends the local model parameters of at least one neuron of the second target layer.

[0076] A ninth aspect of the present application provides a first device, including:

[0077] A processing module, configured to determine a portion of local model parameters of a first model of a first device to be sent, where the portion of local model parameters is obtained by training the first model;

[0078] The transceiver module is used to send the part of the local model parameters and the first information to the second device, where the first information is used to instruct the first device to send the part of the local model parameters.

[0079] A tenth aspect of the present application provides a second device, including:

[0080] a transceiver module, configured to receive partial local model parameters of a first model and first information from a first device, wherein the first information is used to instruct the first device to send the partial local model parameters, where the partial local model parameters are obtained by training the first model;

[0081] A processing module is used to determine the part of local model parameters according to the first information.

[0082] Based on the ninth aspect or the tenth aspect, in a possible implementation, the part of local model parameters includes local weight parameters of the first model.

[0083] Based on the ninth aspect or the tenth aspect, in a possible implementation, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0084] Based on the ninth aspect or the tenth aspect, in a possible implementation method, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

[0085] Based on the ninth aspect or the tenth aspect, in a possible implementation method, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters of the neurons in that layer.

[0086] Based on the ninth aspect or the tenth aspect, in a possible implementation, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device does not send the local model parameters of at least one first target layer neuron; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device sends the local model parameters of at least one second target layer neuron.

[0087] Based on the ninth aspect, in a possible implementation method, the processing module is specifically used to: determine part of the local model parameters based on the local model parameters obtained by the first device for the Rth round of training of the first model, the communication link status of the first device, and at least one of the computing capabilities of the first device, where the part of the local model parameters are obtained by the first device for the R+1th round of training of the first model, where R is an integer greater than or equal to 1.

[0088] In an eleventh aspect, the present application provides a first device, including:

[0089] The transceiver module is configured to receive a portion of first global model parameters of the first model of the first device from the second device; and receive first information from the second device, the first information being used to instruct the second device to send the portion of the first global model parameters.

[0090] The processing module is configured to update the first model according to the first information and part of the first global model parameters to obtain an updated first model.

[0091] A twelfth aspect of the present application provides a second device, including:

[0092] The transceiver module is used to send part of the first global model parameters of the first model of the first device to the first device; send first information to the first device, and the first information is used to instruct the second device to send part of the first global model parameters.

[0093] Based on the eleventh aspect or the twelfth aspect, in a possible implementation, the part of the first global model parameters includes a global weight parameter of the first model.

[0094] Based on the eleventh aspect or the twelfth aspect, in a possible implementation manner, the global weight parameter includes a global weight or a global weight gradient of the first model.

[0095] Based on the eleventh aspect or the twelfth aspect, in a possible implementation method, all first global model parameters of the first model include N first global model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N first global model parameters. The first indication information corresponding to each first global model parameter in the N first global model parameters is used to indicate whether the second device sends the first global model parameter.

[0096] Based on the eleventh aspect or the twelfth aspect, in a possible implementation method, all the first global model parameters of the first model include the first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the first global model parameters of the P layers of neurons. The second indication information corresponding to the first global model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the second device sends the first global model parameters of each layer of neurons.

[0097] Based on the eleventh aspect or the twelfth aspect, in one possible implementation, all first global model parameters of the first model include first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1;

[0098] The first information includes a first identification bit and a layer sequence number of at least one first target layer in the P layer neurons, the first identification bit being used to instruct the second device not to send the first global model parameter of the at least one first target layer neuron; or

[0099] The first information includes a second identification bit and a layer sequence number of at least one second target layer in the P layer neurons, and the second identification bit is used to instruct the second device to send a first global model parameter of at least one second target layer neuron.

[0100] Based on the eleventh aspect or the twelfth aspect, in a possible implementation method, all first global model parameters of the first model include N first global model parameters obtained by the second device by fusing local model parameters of multiple devices in the M+1 round, where N is an integer greater than or equal to 2; the N first global model parameters correspond one-to-one to the N second global model parameters, and the N second global model parameters are obtained by the second device by fusing local model parameters of multiple devices in the M round, where M is an integer greater than or equal to 1; among some first global model parameters, the ratio of the change between each first global model parameter and the second global model parameter corresponding to the first global model parameter to the second global model parameter is greater than the first ratio.

[0101] In a thirteenth aspect, the present application provides a first apparatus, comprising: a processor and a memory. The memory stores a computer program or computer instructions, and the processor is configured to call and execute the computer program or computer instructions stored in the memory, so that the processor implements any one of the implementations of any one of the first, third, and fifth aspects.

[0102] Optionally, the first device further includes a transceiver, and the processor is used to control the transceiver to transmit and receive signals.

[0103] In a fourteenth aspect, the present application provides a second apparatus, comprising: a processor and a memory. The memory stores a computer program or computer instructions, and the processor is configured to call and execute the computer program or computer instructions stored in the memory, so that the processor implements any one of the implementations of the second, fourth, and sixth aspects.

[0104] Optionally, the second device further includes a transceiver, and the processor is used to control the transceiver to transmit and receive signals.

[0105] In a fifteenth aspect, the present application provides a first device, comprising a processor and an interface circuit, wherein the processor is configured to communicate with other devices via the interface circuit and execute the method described in any one of the first, third, and fifth aspects. The processor comprises one or more processors.

[0106] In a sixteenth aspect, the present application provides a second device, comprising a processor and an interface circuit, wherein the processor is configured to communicate with other devices via the interface circuit and execute the method described in any one of the second, fourth, and sixth aspects. The processor comprises one or more processors.

[0107] In a seventeenth aspect, the present application provides a first device, comprising a processor, configured to be connected to a memory and configured to call a program stored in the memory to execute the method described in any of the first, third, and fifth aspects. The memory may be located within or outside the first device. The processor may include one or more processors.

[0108] In an eighteenth aspect, the present application provides a second device, comprising a processor, connected to a memory, configured to call a program stored in the memory to execute the method described in any of the second, fourth, and sixth aspects. The memory may be located within or outside the second device. The processor may include one or more processors.

[0109] In one implementation, the first device of the seventh aspect, the ninth aspect, the eleventh aspect, the thirteenth aspect, and the fifteenth aspect may be a chip (system).

[0110] In one implementation, the second device of the eighth, tenth, twelfth, fourteenth, and sixteenth aspects may be a chip (system).

[0111] In a nineteenth aspect, the present application provides a computer program product comprising instructions, characterized in that when the computer program product is run on a computer, the computer is enabled to execute any one of the implementation methods of any one of the first to sixth aspects.

[0112] The twentieth aspect of the present application provides a computer-readable storage medium comprising computer instructions, which, when executed on a computer, enables the computer to execute any one of the implementation methods of any one of the first to sixth aspects.

[0113] In aspect 21 of the present application, a chip device is provided, comprising a processor for calling a computer program or computer instruction in a memory so that the processor executes any one of the implementation methods of any one of the above-mentioned aspects 1 to 6.

[0114] Optionally, the processor is coupled to the memory via an interface.

[0115] The twenty-second aspect of the present application provides a communication system, which includes the first device as in the seventh aspect and the second device as in the eighth aspect; or, the communication system includes the first device as in the ninth aspect and the second device as in the tenth aspect; or, the communication system includes the first device as in the eleventh aspect and the second device as in the twelfth aspect.

[0116] As can be seen from the above technical solution, a first device receives first information from a second device, the first information being used to instruct the first device whether to transmit each local model parameter of the first model of the first device. The first device determines, based on the first information, a portion of the local model parameters of the first model to be transmitted. These portion of the local model parameters are obtained by training the first model. The first device transmits these portion of the local model parameters of the first model to the second device. Therefore, the first device can determine these portion of the local model parameters of the first model based on the first information and transmit these portion of the local model parameters of the first model. The terminal device does not need to transmit all the local model parameters of the first model. This reduces the signaling overhead of the first device reporting the local model parameters of the first model. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application;

[0118] FIG2 is a schematic diagram of a first embodiment of a communication method according to an embodiment of the present application;

[0119] FIG3 is a schematic diagram of a format of N global model parameters and N first indication information in the same signaling according to an embodiment of the present application;

[0120] FIG4 is a schematic diagram of another format of N global model parameters and N first indication information in the same signaling according to an embodiment of the present application;

[0121] FIG5 is a schematic diagram showing a format of global model parameters of P layers of neurons and P pieces of second indication information in the same signaling in the first model according to an embodiment of the present application;

[0122] FIG6 is a schematic diagram showing another format of global model parameters of P-layer neurons and P second indication information in the same signaling in the first model according to an embodiment of the present application;

[0123] FIG7 is a schematic diagram of a second embodiment of the communication method according to an embodiment of the present application;

[0124] FIG8 is a schematic diagram of a third embodiment of the communication method according to an embodiment of the present application;

[0125] FIG9 is a schematic structural diagram of the first device according to an embodiment of the present application;

[0126] FIG10 is a schematic structural diagram of a second device according to an embodiment of the present application;

[0127] FIG11 is a schematic structural diagram of a terminal device according to an embodiment of the present application;

[0128] FIG12 is a schematic structural diagram of a network device according to an embodiment of the present application. DETAILED DESCRIPTION

[0129] An embodiment of the present application provides a communication method and related devices for reducing the signaling overhead of a first device sending local model parameters of a first model.

[0130] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0131] References to "one embodiment" or "some embodiments" in this application mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0132] In the description of this application, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, "at least one" means one or more, and "plurality" means two or more. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c; a and b; a and c; b and c; or a, b, and c. Among them, a, b, and c can be single or multiple.

[0133] The technical solution of the present application can be applied to cellular communication systems related to the 3rd Generation Partnership Project (3GPP). For example, the fourth generation (4G) communication system, the fifth generation (5G) communication system, and the communication system after the fifth generation communication system. For example, the sixth generation communication system. For example, the fourth generation communication system may include a long term evolution (LTE) communication system. The fifth generation communication system may include a new radio (NR) communication system. The technical solution of the present application can also be applied to wireless fidelity (WiFi) systems, communication systems that support the integration of multiple wireless technologies, device-to-device (D2D) systems, vehicle to everything (V2X) communication systems, etc.

[0134] A possible communication system applicable to the present application is described below with reference to FIG1 .

[0135] Figure 1 is a schematic diagram of a communication system according to an embodiment of the present application. Referring to Figure 1 , the communication system includes a terminal device, an access network, and a core network. The access network includes access network devices, with which terminal devices can communicate. The core network includes core network devices. Terminal devices can communicate with core network devices through access network devices.

[0136] The following introduces the terminal equipment, access network equipment and core network equipment involved in this application.

[0137] In this application, a terminal device is a device with wireless transceiver functions and computing capabilities. The terminal device can perform machine learning training using local data and send relevant information about the model trained by the terminal device to the network device.

[0138] Terminal equipment may refer to user equipment (UE), access terminal, subscriber unit, user station, mobile station, remote station, remote terminal, mobile device, user terminal, wireless communication device, customer premise equipment (CPE), user agent or user device. The terminal device may also be a satellite phone, a cellular phone, a smart phone, a wireless data card, a wireless modem, a machine type communication device, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an on-board device, a car, a communication device carried on a high-altitude aircraft, a wearable device, a drone, a robot, a terminal in D2D, a terminal in V2X, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, or a terminal device in a future communication network, etc., and this application does not impose any restrictions.

[0139] In this application, the access network device has wireless transceiver functions and also has computing capabilities. The access network device is used to communicate with the terminal device. In other words, the access network device can be a device that connects the terminal device to the wireless network. The access network device can be a network node with computing capabilities. For example, the access network device can be an artificial intelligence (AI) node, a computing power node, or an access network node with AI capabilities of the access network. The access network device can fuse models trained by multiple terminal devices and then send them to these terminal devices. This enables joint learning between multiple terminal devices.

[0140] The access network device may be a node in a radio access network. The access network device may be referred to as a base station, and may also be referred to as a radio access network (RAN) node or RAN device. The access network device may be an evolved Node B (eNB or eNodeB) in LTE, or a next generation node B (gNB) in a 5G network, or a base station in a future evolved public land mobile network (PLMN), a broadband network service gateway (BNG), an aggregation switch, or a non-third generation partnership project (3GPP) access device, etc. Optionally, the access network device in the embodiment of the present application may include various forms of base stations. For example, macro base stations, micro base stations (also called small stations), relay stations, access points, devices that implement base station functions in communication systems evolved after 5G, access points (APs) in WiFi systems, transmission points (TRPs), transmitting points (TPs), mobile switching centers, devices that perform base station functions in D2D communications, V2X device communications, or machine-to-machine (M2M) communications. Access network equipment can also include centralized units (CUs) and distributed units (DUs) in cloud radio access network (C-RAN) systems, and access network equipment in non-terrestrial network (NTN) communication systems, that is, they can be deployed on high-altitude platforms or satellites, and this application does not impose any restrictions.

[0141] In this application, the core network device is a control plane network function provided by the network, which is responsible for access control, registration management, service management, mobility management, etc. of terminal devices accessing the network. In the embodiment of this application, the core network device can be the access and mobility management function (AMF) in the 5G communication system, or the core network device in the future network, etc. The core network device can be a network node with computing capabilities. For example, the core network device can be an AI node, a computing power node, or a core network node with AI capabilities of the core network. This application does not limit the specific type of the core network device. In different communication systems, the name of the core network device may be different.

[0142] The communication system to which the technical solution of this application applies includes a first device and a second device. The following describes some possible configurations of the first device and the second device. This application is still applicable to other configurations, and the following examples do not limit this application.

[0143] 1. The first device is a terminal device or a chip in the terminal device, and the second device is a network device or a chip in the network device.

[0144] 2. The first device is an access network device or a chip within the access network device, and the second device is a core network device or a chip within the core network device.

[0145] 3. The first device is a terminal device or a chip in a terminal device, and the second device is a core network device or a chip in a core network device.

[0146] 4. The first device is the first access network device or a chip within the first access network device, and the second device is the first access network device or a chip within the first access network device.

[0147] 5. The first device is a first core network device or a chip within the first core network device, and the second device is a second core network device or a chip within the second core network device.

[0148] 6. The first device is a terminal device or a chip in the terminal device, and the second device is a server or a chip in the server.

[0149] Research into supporting AI capabilities in 5G networks began in Release 16 using the NWDAF network element. The NWDAF element primarily collects and analyzes data at the application layer, providing external services and API calls. In Release 18, research is underway to expand the functionality of the NWDAF element, enabling support for external AI services and model transmission within the network.

[0150] The combination of AI and the network will be an important direction for future research. Model-related parameters need to be transmitted in large quantities across the network. As the scale of the model grows, the number of model-related parameters also increases. Therefore, in wireless networks, the transmission of model-related parameters brings huge signaling overhead. Therefore, how to reduce the signaling overhead of transmitting model-related parameters between devices is a problem worth considering. This application provides a corresponding technical solution for reducing the signaling overhead of the first device or the second device sending model parameters. For details, please refer to the relevant introduction of the embodiments shown in Figures 2, 7, and 8 below.

[0151] The technical solution provided in this application is applicable to a distributed learning communication system. Distributed learning is a learning method for achieving federated learning. Specifically, multiple first devices train local models using local data. A second device then fuses these local models to form a global model. This allows for federated learning while protecting the privacy of user data from multiple first devices. Optionally, distributed learning includes federated learning, split learning, or transfer learning.

[0152] In order to facilitate understanding of the technical solution of this application, the neural network is introduced below.

[0153] A neural network can be composed of neurons, which can be represented by x s The output of the operation unit with the intercept 1 as input can be:

[0154]

[0155] Where, s = 1, 2, ... n, n is a natural number greater than 1, W s is x s The weight of . It should be noted that, optional, x s The weight can also be calculated by adding the weight gradient to the last weight used by the neuron. b is the neuron's bias. f is the neuron's activation function, which is used to introduce nonlinear characteristics into the neural network to convert the input signal in the neuron into the output signal. In other words, inputting an input parameter into a neuron will output the corresponding output parameter. A neural network is a network formed by connecting many of the above-mentioned single neurons together, that is, the output of one neuron can be the input of another neuron.

[0156] Neural networks can have multiple layers of neurons. Below, we'll use a deep neural network (DNN) as an example. A deep neural network is a neural network with many hidden layers. The term "multi-layer neural network" and "deep neural network" are essentially the same thing. Based on the location of different layers within a DNN, the neural network can be divided into three categories: input layer, hidden layer, and output layer. Generally speaking, the first layer is the input layer, the last layer is the output layer, and all layers in between are hidden layers. Layers are fully connected, meaning that any neuron in layer i is connected to any neuron in layer i+1. In deep neural networks, more hidden layers allow the network to better capture complex real-world scenarios. Theoretically, a model with more parameters has higher complexity and a greater "capacity," meaning it can handle more complex learning tasks.

[0157] The technical solution of this application is introduced below in conjunction with specific embodiments.

[0158] FIG2 is a schematic diagram of a first embodiment of a communication method according to an embodiment of the present application. Referring to FIG2 , the method includes:

[0159] 201. A second device sends first information to a first device. The first information is used to indicate whether the first device should send local model parameters of a first model of the first device. In response, the first device receives the first information from the second device.

[0160] Local model parameters refer to model parameters obtained by the first device by training the first model based on the local data of the first device. In other words, model parameters obtained by training the first model using the local data of the first device as input parameters of the first model can be called local model parameters.

[0161] Optionally, the local model parameter is a local weight parameter or other related parameter of the first model, which is not specifically limited in this application. For example, the output parameter of the first model. Optionally, the local weight parameter includes the local weight or local weight gradient of the first model.

[0162] The local model parameters of the first model include all or part of the local model parameters of the first model. The following mainly uses the example that the local model parameters of the first model include all the local model parameters of the first model to introduce the technical solution of the present application.

[0163] Some possible implementations of the first information are introduced below.

[0164] Implementation 1: All local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2. The first information includes N first indication information, each of the N first indication information corresponding to the N local model parameters. The first indication information corresponding to each of the N local model parameters is used to instruct the first apparatus whether to send the local model parameter.

[0165] Optionally, each first indication information in the N first indication information includes one bit, so the N first indication information includes N bits. For example, if the value of one first indication information in the N first indication information is 1, the first indication information is used to instruct the first device to send the local model parameters corresponding to the first indication information. If the value of the one first indication information is 0, the first indication information is used to instruct the first device not to send the local model parameters corresponding to the first indication information. Or, for example, if the value of one first indication information in the N first indication information is 0, the first indication information is used to instruct the first device to send the local model parameters corresponding to the first indication information. If the value of one first indication information is 1, the first indication information is used to instruct the first device not to send the local model parameters corresponding to the first indication information.

[0166] Optionally, the N bits constitute a first bit sequence. For example, the N local model parameters include 10 local model parameters, namely local model parameter 1 to local model parameter 10. The first bit sequence is 1000111001, where the first bit corresponds to local model parameter 1, the second bit corresponds to local model parameter 2, and so on, until the tenth bit corresponds to local model parameter 10. Thus, the second device instructs the first device to send local model parameter 1, local model parameters 5 to local model parameters 7, and local model parameter 10 via the first bit sequence. Other local model parameters do not need to be sent.

[0167] Optionally, the N bits are N elements in the first matrix. The N elements correspond one-to-one to N local model parameters. One of the N elements is used to indicate whether the first device sends the local model parameter corresponding to the element. For example, the first model is a neural network model, and the dimension of the first matrix is ​​determined according to the number of layers included in the neural network model and the number of local model parameters included in each layer of neurons. The neural network model includes 5 layers of neurons, and each layer of neurons includes 4 local model parameters. Therefore, the dimension of the first matrix can be 5*4.

[0168] Optionally, the embodiment shown in Figure 2 further includes step 201a. Step 201a may be performed before step 203a.

[0169] 201a. The second device sends N global model parameters of the first model to the first device. Correspondingly, the first device receives the N global model parameters of the first model from the second device.

[0170] Specifically, the second device fuses the local model parameters of the multiple first devices to obtain N global model parameters of the first model. Then, the second device sends the N global model parameters of the first model to the first device.

[0171] Global model parameters are obtained by the second device by fusing the local model parameters of multiple first devices. Specifically, the second device calculates the global model parameters of the first model based on the local model parameters of multiple first devices and performs corresponding operations. For example, if the first model is a neural network model, multiple first devices may report the local model parameters of neuron 1 within the neural network model. The second device then averages the local model parameters of neuron 1 reported by these multiple first devices to obtain the global model parameters of neuron 1.

[0172] The N global model parameters of the first model correspond one-to-one to the N local model parameters of the first model. For example, the first model is a neural network model. The N global model parameters include eight global model parameters, namely global model parameter 1 to global model parameter 8. The N local model parameters include eight local model parameters, namely local model parameter 1 to local model parameter 8. Global model parameter 1 is the global model parameter of neuron 1. Local model parameter 1 is the local model parameter of neuron 1. Therefore, global model parameter 1 corresponds to local model parameter 1. Similarly, global model parameter 8 is the global model parameter of neuron 8, and local model parameter 8 is the local model parameter of neuron 8. Therefore, global model parameter 8 corresponds to local model parameter 8.

[0173] Two possible ways of sending the N global model parameters and the N first indication information of the first model are described below.

[0174] 1. The N global model parameters of the first model and the N first indication information are carried in the same signaling.

[0175] Specifically, the N first indication information are delivered together with the N global model parameters of the first model. Two possible formats of the N global model parameters of the first model and the N first indication information in the same signaling are introduced below.

[0176] A. N global model parameters and N first indication information are arranged alternately, and the first indication information corresponding to each global model parameter is arranged adjacently after the global model parameter.

[0177] For example, each of the N first indication information includes one bit. As shown in Figure 3, the N global model parameters include eight global model parameters, namely global model parameter 1 to global model parameter 8. The value of global model parameter 1 is 100, and the global model parameter 1 corresponds to the first indication information 1, and the value of the first indication information 1 is 1. That is, the global model parameter 1 is immediately followed by the first indication information 1. The first indication information 1 is used to indicate whether the first device sends the local model parameter 1 corresponding to the first indication information 1. Similarly, the value of the global model parameter 8 is 101, and the global model parameter 8 corresponds to the first indication information 8. The first indication information 8 is used to indicate whether the first device sends the local model parameter 8 corresponding to the first indication information 8.

[0178] B. The N global model parameters are arranged before the N first indication information. That is, the N global model parameters are sent first, and then the N first indication information is sent. It is understandable that the interval between each global model parameter and the first indication information corresponding to the global model parameter is equal.

[0179] For example, each of the N first indication information includes one bit. As shown in Figure 4, the N global model parameters include eight global model parameters, namely global model parameter 1 to global model parameter 8. The eight global model parameters are arranged at intervals. The N first indication information includes eight bits, and the eight bits constitute a first bit sequence. The first bit sequence is arranged after the eight global model parameters. Global model parameter 1 corresponds to the first bit in the first bit sequence, and the first bit is used to indicate whether the first device sends the local model parameter 1 corresponding to the bit. Similarly, the global model parameter 8 corresponds to the eighth bit in the first bit sequence, and the eighth bit is used to indicate whether the first device sends the local model parameter 8 corresponding to the bit.

[0180] Optionally, the N global model parameters of the first model and the N first indication information may be carried in the same radio resource control (RRC) signaling.

[0181] 2. The N global model parameters of the first model and the N first indication information are carried in different signaling.

[0182] In this implementation, the second device sends the N global model parameters and the N first indication information separately.

[0183] For example, each of the N first indication information includes one bit, the N first indication information includes N bits, and the N bits constitute a first bit sequence. The second device separately sends the N global model parameters and the first bit sequence.

[0184] Optionally, the N global model parameters of the first model and the N first indication information may be carried in different RRC signaling.

[0185] Implementation 2: All local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1. The first information includes P second indication information, and the P second indication information correspond one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to instruct the first device whether to send the local model parameters.

[0186] Optionally, each second indication information in the P second indication information includes one bit, so the P second indication information includes P bits. For example, if the value of one second indication information in the P second indication information is 1, the second indication information is used to instruct the first device to send the local model parameters of the neurons of the layer corresponding to the second indication information. If the value of one second indication information is 0, the second indication information is used to instruct the first device not to send the local model parameters of the neurons of the layer corresponding to the second indication information. Alternatively, if the value of one second indication information in the P second indication information is 0, the second indication information is used to instruct the first device to send the local model parameters of the neurons of the layer corresponding to the second indication information. If the value of one second indication information is 1, the second indication information is used to instruct the first device not to send the local model parameters of the neurons of the layer corresponding to the second indication information.

[0187] Optionally, the P bits constitute a second bit sequence. For example, the local model parameters of the P-layer neurons include the local model parameters of the five-layer neurons. The second bit sequence is 10001, where the first bit corresponds to the local model parameters of the first-layer neurons, the second bit corresponds to the local model parameters of the second-layer neurons, and so on. The fifth bit corresponds to the local model parameters of the fifth-layer neurons. It can be seen that the second device instructs the first device to send the local model parameters of the first-layer neurons and the local model parameters of the fifth-layer neurons through the second bit sequence. There is no need to send the local model parameters of the neurons in other layers.

[0188] Optionally, the P bits may be P elements in the second matrix, and the P elements correspond one-to-one to the local model parameters of the P layers of neurons. One of the P elements is used to indicate whether the first device sends the local model parameters of the neurons of the layer corresponding to the element. For example, the first model is a neural network model, and the dimension of the second matrix is ​​determined according to the number of layers included in the neural network model. For example, the neural network model includes 5 layers of neurons, so the dimension of the second matrix is ​​5*1.

[0189] Optionally, the embodiment shown in Figure 2 further includes step 201a. Step 201a may be performed before step 203a.

[0190] 201a. The second device sends global model parameters of P-layer neurons of the first model to the first device. Correspondingly, the first device receives the global model parameters of P-layer neurons of the first model from the second device.

[0191] The global model parameters of the P-layer neurons of the first model correspond one-to-one to the local model parameters of the P-layer neurons of the first model.

[0192] For example, the first model includes two layers of neurons, and each layer of neurons includes four global model parameters. For example, the global model parameters of the first layer of neurons include global model parameters 1 to global model parameters 4. The global model parameters of the second layer of neurons include global model parameters 5 to global model parameters 8. The local model parameters of the first layer of neurons include local model parameters 1 to local model parameters 4. The local model parameters of the second layer of neurons include local model parameters 5 to local model parameters 8. The global model parameters of the first layer of neurons correspond to the local model parameters of the first layer of neurons. The global model parameters of the second layer of neurons correspond to the local model parameters of the second layer of neurons.

[0193] Two possible ways of sending the global model parameters of the P-layer neurons of the first model and the P second indication information are introduced below.

[0194] 1. The global model parameters of the P-layer neurons of the first model and the P second indication information are carried in the same signaling.

[0195] Specifically, the P second indication information is sent along with the global model parameters of the P layer neurons of the first model. The following introduces two possible formats of the global model parameters of the P layer neurons of the first model and the P second indication information in the same signaling.

[0196] The global model parameters of A and P layers of neurons and P second indication information are arranged alternately, and the second indication information corresponding to the global model parameters of each layer of neurons is arranged adjacent to the global model parameters of each layer of neurons.

[0197] For example, each of the P second indication information includes one bit. As shown in Figure 5, the global model parameters of the P-layer neurons include the global model parameters of the two-layer neurons. The global model parameters of the first-layer neurons include global model parameters 1 to global model parameters 4. The global model parameters of the second-layer neurons include global model parameters 5 to global model parameters 8. The global model parameters of the first-layer neurons correspond to the second indication information 1. The value of the second indication information 1 is 1. That is, the global model parameters of the first-layer neurons are immediately followed by the second indication information 1. The global model parameters of the second-layer neurons correspond to the second indication information 2, and the value of the second indication information 2 is 0. That is, the global model parameters of the second-layer neurons are immediately followed by the second indication information 2.

[0198] B. The global model parameters of the P layers of neurons are arranged before the P second indication information. Further optionally, the intervals between the global model parameters of each layer of neurons and the second indication information corresponding to the global model parameters of each layer of neurons are equal.

[0199] For example, each of the P second indication information includes one bit. As shown in Figure 6, the global model parameters of the P layers of neurons include the global model parameters of the two layers of neurons. The global model parameters of the first layer of neurons include global model parameters 1 to global model parameters 4. The global model parameters of the second layer of neurons include global model parameters 5 to global model parameters 8. The global model parameters of the two layers of neurons are arranged at intervals. The P second indication information includes two bits, and the two bits constitute a second bit sequence. The second bit sequence is arranged after the global model parameters of the two layers of neurons. The global model parameters of the first layer of neurons correspond to the first bit in the second bit sequence, and the first bit is used to indicate whether the first device sends the local model parameters of the first layer of neurons corresponding to the bit. The second bit is used to indicate whether the first device sends the local model parameters of the second layer of neurons corresponding to the bit.

[0200] Optionally, the global model parameters of the P layer neurons and the P second indication information can be carried in the same RRC signaling.

[0201] 2. The global model parameters of the P-layer neurons of the first model and the P second indication information are carried in different signalings.

[0202] In this implementation, the second device sends the global model parameters of the P layers of neurons and the P second indication information separately.

[0203] For example, each of the P second indication information includes one bit, and the P second indication information includes P bits. The P bits constitute a second bit sequence. The second device separately sends the global model parameters of the P layer neurons and the second bit sequence.

[0204] Optionally, the global model parameters of the P layer neurons and the P second indication information can be carried in different RRC signaling.

[0205] Implementation method 3: All local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1. The first information includes a first identification bit and the layer sequence number of at least one first target layer in the P-layer neurons. The first identification bit is used to indicate that the first device does not send the local model parameters of the neurons of the at least one first target layer. Alternatively, the first information includes a second identification bit and the layer sequence number of at least one second target layer in the P-layer neurons. The second identification bit is used to indicate that the first device sends the local model parameters of the neurons of the at least one second target layer.

[0206] For example, the P-layer neurons include ten layers of neurons. The layer sequence number of the first layer is 1, the layer sequence number of the second layer is 2, and so on, the layer sequence number of the tenth layer is 10. The first information includes as shown in Table 1, the at least one first target layer includes the third layer and the seventh layer, so the first information includes the layer sequence number of the third layer, the layer sequence number of the seventh layer, and the first identification bit as shown in Table 1. The value of the first identification bit is 0, which is used to indicate that the first device does not send the local model parameters of the neurons in the third layer and the local model parameters of the neurons in the seventh layer.

[0207] Table 1

[0208]

[0209] Therefore, it can be seen that in a scenario where the number of layers of the first target layer is relatively small, the second device can use this implementation method to send the first information, thereby reducing the signaling overhead generated by the second device sending the first information.

[0210] For example, the P-layer neurons include five layers of neurons. The layer sequence number of the first layer is 1, the layer sequence number of the second layer is 2, and so on, the layer sequence number of the fifth layer is 5. The first information includes, as shown in Table 2, the at least one second target layer includes the first layer and the third layer. Therefore, the first information includes the layer sequence number of the first layer, the layer sequence number of the third layer, and the second identification bit as shown in Table 2. The value of the second identification bit is 1, which is used to indicate that the first device sends the local model parameters of the neurons in the first layer and the local model parameters of the neurons in the third layer.

[0211] Table 2

[0212]

[0213] It can be seen from this that, in a scenario where the number of layers of the second target layer is relatively small, the second device can use this implementation method to send the first information, thereby reducing the signaling overhead generated by the second device sending the first information.

[0214] It should be noted that the above-mentioned implementation methods 2 and 3 illustrate implementation methods in which the second device instructs the first device, through the first information, whether to send the local model parameters of each layer of neurons in the P layer of neurons. In actual applications, on this basis, for the layer to be sent, the second device can further instruct the first device which local model parameters of the neurons in the layer to be sent to be sent, which is not limited in this application. For example, the first information also includes third indication information, which is used to instruct the first device whether to send each local model parameter in the neurons in the layer to be sent.

[0215] Optionally, the second device determines the first information based on at least one of the local model parameters reported by multiple first devices, the global model parameters obtained by the second device by integrating the local model parameters reported by multiple first devices, the communication link status of the multiple first devices, and the computing capabilities of the multiple first devices.

[0216] For example, if the communication link states of the multiple first devices are poor or the computing capabilities of the multiple first devices are poor, the second device may instruct the first device to report fewer local model parameters of the first model through the first information.

[0217] For example, if the global model parameter obtained by the second device by fusing the local model parameters reported by multiple first devices in round R+1 has a smaller change than the global model parameter obtained by fusing the local model parameters reported by multiple first devices in round R, the second device may instruct the first device, through the first information, to report the local model parameter corresponding to the global model parameter with the larger change. R is an integer greater than or equal to 1.

[0218] The first device can accurately update the first model in combination with the global model parameters.

[0219] It should be noted that, in this embodiment, the first information determined by the second device for different first devices may be the same or different, and this application does not impose any specific limitation.

[0220] 202. The first device determines part of the local model parameters of the first model to be sent according to the first information.

[0221] The local model parameters are obtained by training the first model.

[0222] For example, as shown in FIG3 , the first device may determine the portion of local model parameters according to eight first indication information. Specifically, the portion of local model parameters includes local model parameter 1, local model parameter 4, local model parameter 6, and local model parameter 8.

[0223] For example, as shown in Figure 5, the first device determines the portion of local model parameters according to the two second indication information. Specifically, the portion of local model parameters includes local model parameters of the first layer of neurons, specifically including local model parameters 1 to local model parameters 4.

[0224] 203. The first device sends the portion of local model parameters to the second device. Correspondingly, the second device receives the portion of local model parameters from the first device.

[0225] Optionally, the embodiment shown in FIG2 further includes step 203a. Step 203a may be performed before step 203.

[0226] 203a. The first device trains the first model to obtain some local model parameters of the first model.

[0227] In this implementation, after the first device determines the partial local model parameters of the first model to be transmitted based on the first information, the first device may only calculate the partial local model parameters of the first model and may not calculate the local model parameters of the first model that do not need to be transmitted. This reduces the amount of local computation performed by the first device and reduces energy consumption.

[0228] Optionally, based on the above step 201a, the embodiment shown in Figure 2 further includes step 201b. Step 201b can be performed before step 203a.

[0229] 201b. The first device updates the first model according to the N global model parameters of the first model or the global model parameters of the P layer of neurons to obtain an updated first model.

[0230] Optionally, based on step 201b, the above step 203a specifically includes: the first device trains the updated first model to obtain some local model parameters of the first model.

[0231] It should be noted that, optionally, the effective time of the first information in the above step 201 may be the time interval between the moment when the second device sends the first information and the moment when the second device updates the first information.

[0232] Optionally, if the second device desires the first device to transmit all local model parameters of the first model, the second device may transmit updated first information to the first device. Optionally, the updated first information may be an all-zero bit sequence, which instructs the first device to transmit all local model parameters of the first model. Alternatively, the first information may be a stop signaling message, which instructs the first device to transmit all local model parameters of the first model.

[0233] For example, when a first condition is met, the second device sends updated first information to the first device. The updated first information is used to instruct the first device to send all local model parameters of the first model. The first condition includes at least one of the following: sufficient computing resources of the first device; sufficient communication resources between the first device and the second device; or idle traffic of the first device.

[0234] It should be noted that there is no fixed order for executing steps 201a to 201b, step 203a, and step 202. Steps 201a to 201b, and step 203a may be executed first, followed by step 202; or step 202 may be executed first, followed by steps 201a to 201b, and step 203a; or, depending on the circumstances, steps 201a to 201b, step 203a, and step 202 may be executed sequentially. This application does not impose any specific restrictions on this order.

[0235] As can be seen, the second device determines the first information based on the factors described above. The first device then sends a portion of the local model parameters of the first model to the first device. The second device accurately determines the global model parameters of the first model based on these partial model parameters and sends the global model parameters of the first model to the first device. The global model parameters of the first model are used by the first device to update the first model. This reduces the overhead of the first device sending the local model parameters of the first model while ensuring the accuracy of the first model.

[0236] In an embodiment of the present application, a first device receives first information from a second device, where the first information is used to indicate whether the first device should send each local model parameter of the first model of the first device. The first device determines, based on the first information, a portion of the local model parameters of the first model to be sent. The portion of the local model parameters is obtained by training the first model. The first device sends the portion of the local model parameters of the first model to the second device. Therefore, it can be seen that the first device can determine the portion of the local model parameters of the first model based on the first information and send the portion of the local model parameters of the first model. The first device does not need to send all the local model parameters of the first model. This reduces the signaling overhead of the first device sending the local model parameters of the first model. This significantly reduces the amount of data transmitted between devices for local model parameter transmission, improves communication efficiency, and reduces the energy consumption generated by transmitting local model parameters between devices, thereby achieving energy saving.

[0237] It should be noted that in the embodiment shown in FIG2 above, steps 201a and 201b illustrate a scheme in which the second device sends N global model parameters of the first model or global model parameters of P-layer neurons to the first device, and the first device updates the first model based on the N global model parameters of the first model or global model parameters of P-layer neurons. In actual applications, the second device may send part of the global model parameters of the first model to the first device, and the first device updates the first model based on the part of the global model parameters. The specific implementation process is similar to the process of steps 801 to 803 in the embodiment shown in FIG8 below. For details, please refer to the relevant introduction of steps 801 to 803 in the embodiment shown in FIG8 below.

[0238] FIG7 is a schematic diagram of a second embodiment of the communication method according to an embodiment of the present application. Referring to FIG7 , the method includes:

[0239] 701. A first device determines partial local model parameters of a first model of the first device to be sent.

[0240] Part of the local model parameters of the first model are obtained by training the first model. For the meaning of the local model parameters, please refer to the above-mentioned related introduction.

[0241] Optionally, the local model parameters include local weight parameters of the first model or other model-related parameters, which are not specifically limited in this application. For example, the output parameters of the first model. Optionally, the local weight parameters of the first model include the local weight or local weight gradient of the first model.

[0242] The following describes a possible implementation method in which the first device determines the partial local model parameters of the first model of the first device to be sent. This application is still applicable to other implementation methods, and this application does not limit them specifically.

[0243] Optionally, step 701 specifically includes: the first device determining, to be transmitted, partial local model parameters of the first model of the first device based on local model parameters obtained by the first device through the Rth round of training of the first model, a communication link status of the first device, and a computing capability of the first device. The partial local model parameters are obtained by the first device through the R+1th round of training of the first model, where R is an integer greater than or equal to 1.

[0244] For example, when the communication link status of the first device is poor, the first device may determine fewer local model parameters of the first model to be sent.

[0245] For example, when the computing capability of the first device is relatively poor, the first device may determine fewer local model parameters of the first model to be sent.

[0246] For example, the first device may determine, among all local model parameters obtained by the first device performing the R+1th round of training on the first model, a local model parameter having a larger change relative to all local model parameters obtained by the first device performing the Rth round of training on the first model. The first device may determine that the portion of local model parameters includes the local model parameter having the larger change.

[0247] 702. The first device sends some local model parameters and first information of the first model to the second device. Correspondingly, the second device receives some local model parameters and first information of the first model from the first device.

[0248] Optionally, the local model parameters of the first model and the first information may be sent simultaneously or separately, which is not limited in this application. That is, the local model parameters of the first model and the first information may be carried in the same signaling or in different signaling.

[0249] Specifically, the first device determines part of the local model parameters of the first model according to the first information, and determines the global model parameters of the first model through the part of the local model parameters.

[0250] The following describes two possible implementations of the first information. This application is still applicable to other implementations and does not limit them.

[0251] Implementation 1: All local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2. The first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each of the N local model parameters is used to instruct the first apparatus whether to send the local model parameter.

[0252] Optionally, each first indication information in the N first indication information includes one bit. Therefore, the N first indication information include N bits. For example, if the value of one first indication information in the N first indication information is 1, the first indication information is used to instruct the first device to send the local model parameters corresponding to the first indication information. If the value of the first indication information is 0, the first indication information is used to instruct the first device not to send the local model parameters corresponding to the first indication information. Alternatively, if the value of one first indication information in the N first indication information is 0, the first indication information is used to instruct the first device to send the local model parameters corresponding to the first indication information. If the value of the first indication information is 1, the first indication information is used to instruct the first device not to send the local model parameters corresponding to the first indication information.

[0253] For example, the N local model parameters include ten local model parameters, namely local model parameter 1 to local model parameter 10. The N bits constitute a first bit sequence, which is 1000100111, where the first bit corresponds to local model parameter 1, the second bit corresponds to local model parameter 2, and so on, with the tenth bit corresponding to local model parameter 10. If the value of a bit in the first bit sequence is 1, it indicates that the first device sends the local model parameter corresponding to the bit. If the value of a bit in the first bit sequence is 0, it indicates that the first device does not send the local model parameter corresponding to the bit. Therefore, the second device can determine that the partial model parameters include local model parameter 1, local model parameter 5, local model parameter 8, local model parameter 9, and local model parameter 10 based on the first bit sequence.

[0254] Optionally, the N bits may be N elements in the first matrix. The N elements correspond one-to-one to the N local model parameters. One of the N elements is used to indicate whether the first device sends the local model parameter corresponding to the element. For example, the first model is a neural network model, and the dimension of the first matrix is ​​determined based on the number of layers included in the neural network model and the number of local model parameters included in each layer of neurons. For example, the neural network model includes 5 layers of neurons, and each layer of neurons includes 4 local model parameters. Therefore, the dimension of the first matrix may be 5*4.

[0255] Implementation 2: All local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1. The first information includes P pieces of second indication information, each of the P pieces of second indication information corresponding one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to instruct the first device whether to send the local model parameters of the neurons in that layer.

[0256] Optionally, each second indication information in the P second indication information includes one bit, so the P second indication information includes P bits. For example, if the value of one second indication information in the P second indication information is 1, the second indication information is used to instruct the first device to send the local model parameters of the neurons of the layer corresponding to the second indication information. If the value of the second indication information is 0, the second indication information is used to instruct the first device not to send the local model parameters of the neurons of the layer corresponding to the second indication information. Alternatively, if the value of one second indication information in the P second indication information is 0, the second indication information is used to instruct the first device to send the local model parameters of the neurons of the layer corresponding to the second indication information. If the value of one second indication information is 1, the second indication information is used to instruct the first device not to send the local model parameters of the neurons of the layer corresponding to the second indication information.

[0257] For example, the local model parameters of the P-layer neurons include the local model parameters of the five-layer neurons. The P bits constitute a second bit sequence. The second bit sequence is 10010, where the first bit corresponds to the local model parameters of the first-layer neurons, the second bit corresponds to the local model parameters of the second-layer neurons, and so on, the fifth bit corresponds to the local model parameters of the fifth-layer neurons. If the value of a bit in the second bit sequence is 1, it indicates that the first device sends the local model parameters of the neurons of the layer corresponding to the bit. If the value of a bit in the second bit sequence is 0, it indicates that the first device does not send the local model parameters of the neurons of the layer corresponding to the bit. It can be seen that the second device can determine that the part of the model parameters includes the local model parameters of the first-layer neurons and the local model parameters of the fourth-layer neurons based on the second bit sequence.

[0258] Optionally, the P bits may be P elements in the second matrix, and the P elements correspond one-to-one to the local model parameters of the P layers of neurons. One of the P elements is used to indicate whether the first device sends the local model parameters of the neurons of the layer corresponding to the element. For example, the first model is a neural network model, and the dimension of the second matrix is ​​determined according to the number of layers included in the neural network model. For example, the neural network model includes 5 layers of neurons, so the dimension of the second matrix is ​​5*1.

[0259] 3. All local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device does not send the local model parameters of at least one neuron of the first target layer; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device sends the local model parameters of at least one neuron of the second target layer.

[0260] For the relevant example introduction of the implementation method 3, please refer to the relevant introduction in Table 1 and Table 2 in the embodiment shown in Figure 2 above, which will not be repeated here.

[0261] It should be noted that the above-mentioned implementation methods 2 and 3 illustrate implementation methods in which the first device instructs the first device via the first information whether to send the local model parameters of the neurons in each layer of the P layer of neurons. In this actual application, on this basis, for the layer to be sent, the second device can further instruct the first device which local model parameters of the neurons in the layer to be sent to be sent, and this application does not limit this. For example, the first information also includes third indication information, and the third indication information is used to instruct the first device whether to send the local model parameters of the neurons in the layer to be sent.

[0262] Optionally, the embodiment shown in FIG7 further includes step 702a. Step 702a may be performed before step 702.

[0263] 702a. The first device trains the first model to obtain some local model parameters of the first model.

[0264] In this implementation, the first device determines a portion of local model parameters of the first model to be transmitted. The first device only calculates the portion of local model parameters of the first model. The first device may not calculate the local model parameters of the first model that do not need to be transmitted. This reduces the amount of local computation performed by the first device and reduces energy consumption.

[0265] Optionally, the embodiment shown in Figure 7 further includes step 701a and step 701b. Step 701a and step 701b can be performed before step 701a.

[0266] 701a. The second device sends N global model parameters of the first model to the first device. Correspondingly, the first device receives the N global model parameters of the first model from the second device.

[0267] 701b. The first device updates the first model according to the N global model parameters of the first model to obtain an updated first model.

[0268] For the meaning of global model parameters, please refer to the above introduction.

[0269] Optionally, based on the above steps 701a and 701b, the above step 702a specifically includes: the first device trains the updated first model to obtain some local model parameters of the first model.

[0270] It should be noted that there is no fixed order for executing steps 701a to 701b and step 702a, and step 701. Steps 701a to 701b and step 702a may be executed first, followed by step 702a; or step 702a may be executed first, followed by steps 701a to 701b and step 702a; or, depending on the circumstances, steps 701a to 701b, step 702a, and step 701 may be executed simultaneously, and this application does not impose any specific restrictions.

[0271] In an embodiment of the present application, a first device determines partial local model parameters of a first model of the first device to be transmitted; then, the first device transmits the partial local model parameters of the first model and first information to a second device. The first information is used to instruct the first device to transmit the partial local model parameters of the first model. Therefore, it can be seen that the first device can only transmit the partial local model parameters of the first model, and the first device does not need to transmit all the local model parameters of the first model. This reduces the signaling overhead of the first device transmitting the local model parameters of the first model. This significantly reduces the amount of data transmitted between devices for local model parameter transmission, improves communication efficiency, and reduces the energy consumption generated by transmitting local model parameters between devices, thereby achieving energy conservation.

[0272] It should be noted that steps 701a and 701b of the embodiment shown in FIG. 7 illustrate a scheme in which the second device sends N global model parameters of the first model to the first device, and the first device updates the first model based on the N global model parameters of the first model. In actual applications, the second device may send a portion of the global model parameters of the first model to the first device, and the first device may update the first model based on the portion of the global model parameters. The specific implementation process is similar to steps 801 to 803 in the embodiment shown in FIG. 8 below. For details, please refer to the relevant description of steps 801 to 803 in the embodiment shown in FIG. 8 below.

[0273] FIG8 is a schematic diagram of a third embodiment of the communication method according to an embodiment of the present application. Referring to FIG8 , the method includes:

[0274] 801. A second device sends a portion of first global model parameters of a first model of the first device to a first device. Correspondingly, the first device receives a portion of the first global model parameters of the first model of the first device from the second device.

[0275] For the meaning of global model parameters, please refer to the above introduction.

[0276] For example, during federated learning, the second device may derive the first global model parameters of the first model of the first device by fusing the local model parameters of multiple first devices. The second device may then select some of the first global model parameters of the first model and send these to the first device.

[0277] Optionally, all first global model parameters of the first model include N first global model parameters obtained by the second device in the M+1th round by fusing local model parameters of multiple devices. N is an integer greater than or equal to 2. The N first global model parameters correspond one-to-one to the N second global model parameters, and the N second global model parameters are obtained by the second device in the Mth round by fusing the local model parameters of multiple devices, where M is an integer greater than or equal to 1. For some first global model parameters of the first model, the ratio of a change between each first global model parameter and the second global model parameter corresponding to the first global model parameter to the second global model parameter is greater than the first ratio.

[0278] That is, for each of the N first global model parameters, if the amount of change in the first global model parameter relative to the second global model parameter corresponding to the first global model parameter is large, the second device may send the first global model parameter to the first device. If the amount of change in the first global model parameter relative to the second global model parameter corresponding to the first global model parameter is small, the second device may not send the first global model parameter.

[0279] Optionally, the first ratio may be 1 / 10 or 1 / 15, which is not specifically limited in this application.

[0280] Optionally, the first ratio may be set based on at least one of the size of the data sample, the type of the first model, and the capacity of the first model. The data sample refers to a plurality of local model parameters of the first device collected by the second device. For example, the larger the capacity of the first model and the more complex the first model, the smaller the first ratio may be. For example, if the data sample is relatively sufficient, the first ratio may be set to a larger value.

[0281] 802. The second device sends first information to the first device. The first information is used to instruct the second device to send part of the first global model parameters of the first model. Correspondingly, the first device receives the first information from the second device.

[0282] The following describes two possible implementations of the first information. This application is still applicable to other implementations, and this application does not limit them specifically.

[0283] 1. All first global model parameters of the first model include N first global model parameters, where N is an integer greater than or equal to 2. The first information includes N first indication information, and the N first indication information corresponds one-to-one to the N first global model parameters. The first indication information corresponding to each of the N first global model parameters is used to instruct the second apparatus whether to send the first global model parameter.

[0284] Optionally, each first indication information in the N first indication information includes one bit. Therefore, the N first indication information include N bits. For example, if the value of one first indication information in the N first indication information is 1, the first indication information is used to instruct the second device to send the first global model parameter corresponding to the first indication information. If the value of one first indication information in the N first indication information is 0, the first indication information is used to instruct the second device to send the first global model parameter corresponding to the first indication information. Alternatively, if the value of one first indication information in the N first indication information is 0, the first indication information is used to instruct the second device to send the first global model parameter corresponding to the first indication information. If the value of one first indication information in the N first indication information is 1, the first indication information is used to instruct the second device to send the first global model parameter corresponding to the first indication information.

[0285] For example, the N first global model parameters include ten first global model parameters, namely, first global model parameter 1 to first global model parameter 10. The N bits constitute a first bit sequence, and the first bit sequence is 0111001100, wherein the first bit corresponds to the first global model parameter 1, the second bit corresponds to the first global model parameter 2, and so on, the tenth bit corresponds to the first global model parameter 10. If the value of a bit in the first bit sequence is 1, it instructs the second device to send the first global model parameter corresponding to the bit. If the value of a bit in the first bit sequence is 0, it instructs the second device to send the first global model parameter corresponding to the bit. It can be seen that the first device can determine that the part of the first global model parameters includes the first global model parameters 2 to the first global model parameters 4, the first global model parameter 7, and the first global model parameter 8 based on the first bit sequence.

[0286] Optionally, the N bits may be N elements in the first matrix. The N elements correspond one-to-one to the N first global model parameters. One of the N elements is used to indicate whether the second device sends the first global model parameter corresponding to the element. For example, the first model is a neural network model, and the dimension of the first matrix is ​​determined according to the number of layers included in the neural network model and the number of local model parameters included in each layer of neurons. The neural network model includes 5 layers of neurons, and each layer of neurons includes 4 local model parameters. Therefore, the dimension of the first matrix can be 5*4.

[0287] 2. All first global model parameters of the first model include the first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information correspond one-to-one to the first global model parameters of the P layers of neurons. The second indication information corresponding to the first global model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the second device sends the first global model parameters of each layer of neurons.

[0288] Optionally, each second indication information in the P second indication information includes one bit, so the P second indication information includes P bits. For example, if the value of one second indication information in the P second indication information is 1, then the second indication information is used to instruct the second device to send the first global model parameters of the neurons of the layer corresponding to the second indication information. If the value of one second indication information in the P second indication information is 0, then the second indication information is used to instruct the second device not to send the first global model parameters of the neurons of the layer corresponding to the second indication information. Alternatively, if the value of one second indication information in the P second indication information is 0, then the second indication information is used to instruct the second device to send the first global model parameters of the neurons of the layer corresponding to the second indication information. If the value of one second indication information in the P second indication information is 1, then the second indication information is used to instruct the second device not to send the first global model parameters of the neurons of the layer corresponding to the second indication information.

[0289] For example, the first global model parameters of the P-layer neurons include the first global model parameters of the five-layer neurons. The P bits constitute a second bit sequence. The second bit sequence is 01110. The first bit corresponds to the first global model parameter of the first-layer neurons, the second bit corresponds to the first global model parameter of the second-layer neurons, and so on, the fifth bit corresponds to the first global model parameter of the fifth-layer neurons. If the value of a bit in the second bit sequence is 1, it indicates that the second device sends the first global model parameters of the neurons of the layer corresponding to the bit. If the value of a bit in the second bit sequence is 0, it indicates that the second device does not send the first global model parameters of the neurons of the layer corresponding to the bit. It can be seen that the first device can determine that the part of the first global model parameters includes the first global model parameters of the second-layer neurons, the first global model parameters of the third-layer neurons, and the first global model parameters of the fourth-layer neurons based on the second bit sequence.

[0290] Optionally, the P bits may be P elements in the second matrix, and the P elements correspond one-to-one to the first global model parameters of the neurons in the P layers. One of the P elements is used to indicate whether the second device sends the first global model parameters of the neurons in the layer corresponding to the element. For example, the first model is a neural network model, and the dimension of the second matrix is ​​determined according to the number of layers included in the neural network model. For example, the neural network model includes 5 layers of neurons, so the dimension of the second matrix is ​​5*1.

[0291] Implementation method 3: All first global model parameters of the first model include the first global model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the second device does not send the first global model parameters of at least one first target layer neuron; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the second device sends the first global model parameters of at least one second target layer neuron.

[0292] For example, the P-layer neurons include eight layers of neurons. The layer sequence number of the first layer is 1, the layer sequence number of the second layer is 2, and so on. The layer sequence number of the eighth layer is 8. The first information includes as shown in Table 3. The at least one first target layer includes the second layer and the fourth layer. Therefore, the first information includes the layer sequence number of the second layer and the layer sequence number of the fourth layer as shown in Table 3. The value of the first identification bit is 0, which is used to indicate that the first device does not send the first global model parameters of the neurons in the second layer and the first global model parameters of the neurons in the fourth layer.

[0293] Table 3

[0294] The first identification bit of the layer number is 20

[0295] 4

[0296] Therefore, it can be seen that in a scenario where the number of layers of the first target layer is relatively small, the second device can use this implementation method to send the first information, thereby reducing the signaling overhead generated by the second device sending the first information.

[0297] For example, the P layer of neurons includes five layers of neurons. The layer number of the first layer is 1, the layer number of the second layer is 2, and so on, the layer number of the fifth layer is 5. The first information includes the at least one second target layer as shown in Table 4, including the third layer and the fourth layer. Therefore, the first information includes the layer number of the third layer, the layer number of the fourth layer, and the second identification bit as shown in Table 4. The value of the second identification bit is 1, which is used to indicate that the second device sends the first global model parameters of the neurons in the third layer and the first global model parameters of the neurons in the fourth layer.

[0298] Table 4

[0299]

[0300] It can be seen from this that, in a scenario where the number of layers of the second target layer is relatively small, the second device can use this implementation method to send the first information, thereby reducing the signaling overhead generated by the second device sending the first information.

[0301] It should be noted that the above-mentioned implementation method 2 and implementation method 3 show the implementation method in which the second device instructs the first device through the first information whether to send the first global model parameters of each layer of neurons in the P layer of neurons. That is, the first information is used to indicate which layers of neurons need to send the first global model parameters. In actual applications, on this basis, the second device can further instruct the second device to send which first global model parameters of the neurons in the layer that need to be sent as indicated by the first information, and this application does not limit this specifically. For example, the first information also includes third indication information, and the third indication information is used to instruct the second device whether to send each first global model parameter in the neurons of the layer to be sent.

[0302] In this implementation, different first devices use the same first information.

[0303] It should be noted that, optionally, there is no fixed execution order between the above-mentioned step 801 and the above-mentioned step 802. Step 801 can be executed first, and then step 802; or, step 802 can be executed first, and then step 801; or, step 801 and step 802 can be executed simultaneously depending on the situation, and this application does not make any specific limitations.

[0304] It should be noted that, optionally, the part of the first global model and the first information can be carried in the same signaling or in different signaling.

[0305] 803. The first device updates the first model according to the first information and part of the first global model parameters to obtain an updated first model.

[0306] For example, the first information includes a first bit sequence, which is 0111001100, where the first bit corresponds to the first global model parameter 1, the second bit corresponds to the first global model parameter 2, and so on, the tenth bit corresponds to the first global model parameter 10. If the value of a bit in the first bit sequence is 1, it instructs the second device to send the first global model parameter corresponding to the bit. If the value of a bit in the first bit sequence is 0, it instructs the second device to send the first global model parameter corresponding to the bit. It can be seen that the first device can determine that the part of the first global model parameters includes the first global model parameters 2 to the first global model parameters 4, the first global model parameter 7, and the first global model parameter 8 based on the first bit sequence. The first global model parameters 2 to the first global model parameters 4 correspond to neuron 1, neuron 2, and neuron 3, respectively. The first global model parameter 7 corresponds to neuron 7, and the first global model parameter corresponds to neuron 8. Therefore, the first device can use the first global model parameter 2 as the global model parameter of neuron 1, the first global model parameter 3 as the global model parameter of neuron 2, the first global model parameter 4 as the global model parameter of neuron 3, the first global model parameter 7 as the global model parameter of neuron 7, and the first global model parameter 8 as the global model parameter of neuron 8.

[0307] Optionally, the embodiment shown in FIG8 further includes step 804 and step 805. Step 804 and step 805 may be performed after step 803.

[0308] 804. The first device trains the updated first model to obtain local model parameters.

[0309] 805. The first device sends the local model parameters of the first model to the second device. Correspondingly, the second device receives the local model parameters of the first model from the first device.

[0310] In an embodiment of the present application, a first device receives part of the first global model parameters of the first model of the first device from a second device; the first device receives first information from the second device, and the first information is used to instruct the second device to send the part of the first global model parameters. Then, the first device updates the first model according to the first information and the part of the first global model parameters to obtain an updated first model. It can be seen from this that the second device can only send part of the first global model parameters of the first model to the first device, without having to send all the first global model parameters of the first model. This reduces the signaling overhead of the second device sending the first global model parameters of the first model. That is, the amount of data transmitted between devices for global model parameter transmission is greatly reduced, communication efficiency is improved, and the energy consumption generated by transmitting global model parameters between devices is reduced, thereby achieving energy saving effects.

[0311] It should be noted that steps 804 to 805 in the embodiment shown in FIG. 8 illustrate a scheme in which the first device trains the first model and transmits the local model parameters of the first model to the second device. In practical applications, the first device may only transmit the local model parameters of the first model to the second device, thereby reducing the overhead of transmitting the local model parameters for the first device. For example, the first device may receive information from the second device indicating whether the first device should transmit the local model parameters of the first model. Based on this information, the first device determines the portion of the local model parameters of the first model to be transmitted and transmits these portion of the local model parameters to the second device. This implementation process is similar to steps 201 to 203 in the embodiment shown in FIG. For details, please refer to the description of steps 201 to 203 in the embodiment shown in FIG. 2 . For another example, the first device may independently determine the portion of the local model parameters of the first model to be transmitted. The first device then transmits these portion of the local model parameters and the information instructing the first device to transmit these portion of the local model parameters to the second device. This implementation process is similar to steps 701 to 702 in the embodiment shown in FIG. 7 . For details, please refer to the relevant introduction of steps 701 to 702 in the embodiment shown in FIG. 7 .

[0312] The following describes the first apparatus provided in the embodiments of the present application. Please refer to Figure 9, which is a schematic diagram of the structure of the first apparatus in the embodiments of the present application. First apparatus 900 can be used to perform the steps performed by the first apparatus in the embodiments shown in Figures 2, 7, and 8. For details, please refer to the relevant description of the above method embodiments.

[0313] The first device 900 includes a transceiver module 901 and a processing module 902 .

[0314] In one possible implementation, the first device 900 specifically performs the following solution:

[0315] The transceiver module 901 is configured to receive first information from a second device, where the first information is configured to indicate whether the first device 900 should transmit local model parameters of the first model of the first device 900.

[0316] A processing module 902 is configured to determine, based on the first information, some local model parameters of the first model to be sent, where the some local model parameters are obtained by training the first model;

[0317] The transceiver module 901 is further configured to send the portion of local model parameters to the second device.

[0318] Optionally, the local model parameters include local weight parameters of the first model.

[0319] Optionally, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0320] Optionally, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device 900 sends the local model parameter.

[0321] Optionally, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information correspond one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device 900 sends the local model parameters.

[0322] Optionally, the transceiver module 901 is further configured to receive N global model parameters of the first model or global model parameters of P layers of neurons of the first model from a second device.

[0323] Optionally, N global model parameters correspond one-to-one to N local model parameters; the N first indication information and the N global model parameters are carried in the same signaling or different signaling. When the N first indication information and the N global model parameters are carried in the same signaling, the N global model parameters and the N first indication information are arranged at intervals, and the first indication information corresponding to each global model parameter is arranged adjacent to the global model parameter, or the N global model parameters are arranged before the N first indication information.

[0324] Optionally, the global model parameters of P layer neurons correspond one-to-one to the local model parameters of P layer neurons; the P second indication information and the global model parameters of P layer neurons are carried in the same signaling or different signaling. When the P second indication information and the global model parameters of P layer neurons are carried in the same signaling, the global model parameters of P layer neurons and the P second indication information are arranged at intervals, and the second indication information corresponding to the global model parameters of each layer of neurons is arranged adjacent to the global model parameters of each layer of neurons, or the global model parameters of P layer neurons are arranged before the P second indication information.

[0325] Optionally, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device 900 does not send the local model parameters of at least one first target layer neuron; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device 900 sends the local model parameters of at least one second target layer neuron.

[0326] In another possible implementation, the first device 900 is specifically configured to perform the following solution:

[0327] A processing module 902 is configured to determine a portion of local model parameters of a first model of the first device 900 to be sent, where the portion of local model parameters is obtained by training the first model;

[0328] The transceiver module 901 is used to send the part of local model parameters and first information to the second device, where the first information is used to instruct the first device 900 to send the part of local model parameters.

[0329] Optionally, the part of local model parameters includes local weight parameters of the first model.

[0330] Optionally, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0331] Optionally, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information correspond one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device 900 sends the local model parameter.

[0332] Optionally, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device 900 sends the local model parameters of the neurons in that layer.

[0333] Optionally, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device 900 does not send the local model parameters of at least one first target layer neuron; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device 900 sends the local model parameters of at least one second target layer neuron.

[0334] Optionally, the processing module 902 is specifically used to determine part of the local model parameters based on the local model parameters obtained by the first device 900 through the Rth round of training of the first model, the communication link status of the first device 900, and at least one of the computing capabilities of the first device 900, where the part of the local model parameters are obtained by the first device 900 through the R+1th round of training of the first model, where R is an integer greater than or equal to 1.

[0335] In another possible implementation, the first device 900 is specifically configured to perform the following solution:

[0336] The transceiver module 901 is configured to receive a portion of first global model parameters of a first model of the first device 900 from a second device; and receive first information from the second device, the first information being used to instruct the second device to send a portion of the first global model parameters.

[0337] The processing module 902 is configured to update the first model according to the first information and part of the first global model parameters to obtain an updated first model.

[0338] Optionally, the part of the first global model parameters includes global weight parameters of the first model.

[0339] Optionally, the global weight parameter includes the global weight or global weight gradient of the first model.

[0340] Optionally, all first global model parameters of the first model include N first global model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, the N first indication information correspond one-to-one to the N first global model parameters, and the first indication information corresponding to each first global model parameter in the N first global model parameters is used to indicate whether the second device sends the first global model parameter.

[0341] Optionally, all first global model parameters of the first model include the first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, the P second indication information correspond one-to-one to the first global model parameters of the P layers of neurons, and the second indication information corresponding to the first global model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the second device sends the first global model parameters of each layer of neurons.

[0342] Optionally, all first global model parameters of the first model include first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1;

[0343] The first information includes a first identification bit and a layer sequence number of at least one first target layer in the P layer neurons, the first identification bit being used to instruct the second device not to send the first global model parameter of the at least one first target layer neuron; or

[0344] The first information includes a second identification bit and a layer sequence number of at least one second target layer in the P layer neurons, and the second identification bit is used to instruct the second device to send a first global model parameter of at least one second target layer neuron.

[0345] Optionally, all first global model parameters of the first model include N first global model parameters obtained by the second device in the M+1 round by fusing local model parameters of multiple devices, where N is an integer greater than or equal to 2; the N first global model parameters correspond one-to-one to the N second global model parameters, and the N second global model parameters are obtained by the second device in the M round by fusing local model parameters of multiple devices, where M is an integer greater than or equal to 1; among some first global model parameters, the ratio of the change between each first global model parameter and the second global model parameter corresponding to the first global model parameter to the second global model parameter is greater than the first ratio.

[0346] The following describes the second device provided in the embodiments of the present application. Please refer to Figure 10, which is a schematic diagram of the structure of the second device in the embodiments of the present application. Second device 1000 can be used to perform the steps performed by the second device in the embodiments shown in Figures 2, 7, and 8. For details, please refer to the relevant description of the above method embodiments.

[0347] The second device 1000 includes a transceiver module 1001. Optionally, the second device 1000 further includes a processing module 1002.

[0348] In one possible implementation, the second device 1000 is configured to perform the following solution:

[0349] The transceiver module 1001 is used to send first information to the first device, where the first information is used to indicate whether the first device sends the local model parameters of the first model of the first device; and receive some local model parameters of the first model from the first device, where the some local model parameters are obtained by training the first model.

[0350] Optionally, the local model parameters include local weight parameters of the first model.

[0351] Optionally, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0352] Optionally, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, the N first indication information correspond one-to-one to the N local model parameters, and the first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

[0353] Optionally, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, the P second indication information correspond one-to-one to the local model parameters of the P layers of neurons, and the second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters.

[0354] Optionally, the transceiver module 1001 is further used to send N global model parameters of the first model or global model parameters of P layers of neurons of the first model to the first device.

[0355] Optionally, N global model parameters correspond one-to-one to N local model parameters; the N first indication information and the N global model parameters are carried in the same signaling or different signaling. When the N first indication information and the N global model parameters are carried in the same signaling, the N global model parameters and the N first indication information are arranged at intervals, and the first indication information corresponding to each global model parameter is arranged adjacent to the global model parameter, or the N global model parameters are arranged before the N first indication information.

[0356] Optionally, the global model parameters of P layer neurons correspond one-to-one to the local model parameters of P layer neurons; the P second indication information and the global model parameters of P layer neurons are carried in the same signaling or different signaling. When the P second indication information and the global model parameters of P layer neurons are carried in the same signaling, the global model parameters of P layer neurons and the P second indication information are arranged at intervals, and the second indication information corresponding to the global model parameters of each layer of neurons is arranged adjacent to the global model parameters of each layer of neurons, or the global model parameters of P layer neurons are arranged before the P second indication information.

[0357] Optionally, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device does not send the local model parameters of at least one neuron of the first target layer; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device sends the local model parameters of at least one neuron of the second target layer.

[0358] In another possible implementation, the second device 1000 is configured to perform the following solution:

[0359] The transceiver module 1001 is configured to receive partial local model parameters of a first model and first information from a first device, where the first information is used to instruct the first device to send the partial local model parameters, where the partial local model parameters are obtained by training the first model;

[0360] The processing module 1002 is configured to determine the local model parameters according to the first information.

[0361] Optionally, the part of local model parameters includes local weight parameters of the first model.

[0362] Optionally, the local weight parameter includes a local weight or a local weight gradient of the first model.

[0363] Optionally, all local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, the N first indication information correspond one-to-one to the N local model parameters, and the first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

[0364] Optionally, all local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters of the neurons in that layer.

[0365] Optionally, all local model parameters of the first model include local model parameters of P-layer neurons, where P is an integer greater than or equal to 1; the first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, and the first identification bit is used to indicate that the first device does not send the local model parameters of at least one neuron of the first target layer; or, the first information includes a second identification bit and a layer sequence number of at least one second target layer in the P-layer neurons, and the second identification bit is used to indicate that the first device sends the local model parameters of at least one neuron of the second target layer.

[0366] In another possible implementation, the second device 1000 is configured to perform the following solution:

[0367] The transceiver module 1001 is used to send part of the first global model parameters of the first model of the first device to the first device; send first information to the first device, where the first information is used to instruct the second device 1000 to send part of the first global model parameters.

[0368] Optionally, the part of the first global model parameters includes global weight parameters of the first model.

[0369] Optionally, the global weight parameter includes the global weight or global weight gradient of the first model.

[0370] Optionally, all first global model parameters of the first model include N first global model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, the N first indication information correspond one-to-one to the N first global model parameters, and the first indication information corresponding to each first global model parameter in the N first global model parameters is used to indicate whether the second device 1000 sends the first global model parameter.

[0371] Optionally, all first global model parameters of the first model include the first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, the P second indication information correspond one-to-one to the first global model parameters of the P layers of neurons, and the second indication information corresponding to the first global model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the second device 1000 sends the first global model parameters of each layer of neurons.

[0372] Optionally, all first global model parameters of the first model include first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1;

[0373] The first information includes a first identification bit and a layer sequence number of at least one first target layer in the P layer neurons, the first identification bit being used to instruct the second device 1000 not to send the first global model parameter of the at least one first target layer neuron; or

[0374] The first information includes a second identification bit and a layer sequence number of at least one second target layer in the P layer neurons, and the second identification bit is used to instruct the second device 1000 to send the first global model parameters of the neurons of the at least one second target layer.

[0375] Optionally, all first global model parameters of the first model include N first global model parameters obtained by the second device 1000 in the M+1 round by fusing local model parameters of multiple devices, where N is an integer greater than or equal to 2; the N first global model parameters correspond one-to-one to the N second global model parameters, and the N second global model parameters are obtained by the second device 1000 in the M round by fusing local model parameters of multiple devices, where M is an integer greater than or equal to 1; among some first global model parameters, the ratio of the change between each first global model parameter and the second global model parameter corresponding to the first global model parameter to the second global model parameter is greater than the first ratio.

[0376] The present application also provides a terminal device. Figure 11 is a schematic diagram of the structure of a terminal device 1100 provided in an embodiment of the present application. Terminal device 1100 can be used in the system shown in Figure 1. For example, terminal device 1100 can be the terminal device in the system shown in Figure 1, configured to perform the functions of the first device in the above method embodiment.

[0377] As shown in the figure, the terminal device 1100 includes a processor 1110 and a transceiver 1120. Optionally, the terminal device 1100 also includes a memory 1130. The processor 1110, the transceiver 1120, and the memory 1130 can communicate with each other via internal connection paths to transmit control and / or data signals. The memory 1130 is used to store computer programs, and the processor 1110 is used to call and execute the computer programs from the memory 1130 to control the transceiver 1120 to transmit and receive signals. Optionally, the terminal device 1100 may also include an antenna 1140 for transmitting uplink data or uplink control signaling output by the transceiver 1120 via wireless signals.

[0378] The processor 1110 and the memory 1130 may be combined into a processing device, and the processor 1110 is configured to execute program code stored in the memory 1130 to implement the aforementioned functions. In a specific implementation, the memory 1130 may also be integrated into the processor 1110 or independent of the processor 1110. For example, the processor 1110 may correspond to the processing module 902 in FIG. 9 .

[0379] The transceiver 1120 may correspond to the transceiver module 901 in FIG9 . The transceiver 1120 may also be referred to as a transceiver unit. The transceiver 1120 may include a receiver (or receiver, receiving circuit) and a transmitter (or transmitter, transmitting circuit). The receiver is used to receive signals, and the transmitter is used to transmit signals.

[0380] It should be understood that the terminal device 1100 shown in FIG11 is capable of implementing the various processes involving the first apparatus in the method embodiments shown in FIG2 , FIG7 , and FIG8 . The operations and / or functions of the various modules in the terminal device 1100 are respectively for implementing the corresponding processes in the aforementioned apparatus embodiments. For details, please refer to the description of the aforementioned apparatus embodiments; to avoid repetition, detailed descriptions are omitted here.

[0381] The processor 1110 can be used to execute the actions implemented by the first device described in the previous device embodiment, and the transceiver 1120 can be used to execute the transceiver actions of the first device described in the previous device embodiment. For details, please refer to the description in the previous device embodiment, which will not be repeated here.

[0382] Optionally, the terminal device 1100 may further include a power supply 1150 for providing power to various devices or circuits in the terminal device.

[0383] In addition, in order to make the functions of the terminal device more complete, the terminal device 1100 may also include one or more of an input unit 1160, a display unit 1170, an audio circuit 1180, a camera 1190 and a sensor 1100, and the audio circuit may also include a speaker 1182, a microphone 1184, etc.

[0384] The present application also provides a network device. Please refer to Figure 12, which is a schematic diagram of the structure of a network device 1200 provided in an embodiment of the present application. This network device 1200 can be applied to the system shown in Figure 1. For example, network device 1200 can be an access network device or a core network device in the system shown in Figure 1, and is used to perform the functions of the second device in the above method embodiment. It should be understood that the following is merely an example, and in future communication systems, network devices may have other forms and structures.

[0385] For example, in a 5G communication system, the network device 1200 may include a CU, a DU, and an AAU. Compared to a network device in an LTE communication system, which consists of one or more radio frequency units, such as a remote radio unit (RRU) and one or more baseband units (BBU):

[0386] The non-real-time portion of the original BBU will be separated and redefined as a CU, responsible for handling non-real-time protocols and services. Some of the BBU's physical layer processing functions will be merged with the original RRU and passive antennas into the AAU. The remaining BBU functions will be redefined as a DU, responsible for handling physical layer protocols and real-time services. In short, the CU and DU are differentiated by the real-time nature of their processing, and the AAU is a combination of the RRU and antenna.

[0387] The CU, DU, and AAU can be deployed separately or together, resulting in a variety of network deployment configurations. One possible deployment configuration, shown in Figure 12, is consistent with traditional 4G network equipment, with the CU and DU deployed on shared hardware. It should be understood that Figure 12 is merely an example and does not limit the scope of protection of this application. For example, the deployment configuration may also include the DU being deployed in the BBU room, the CU being deployed centrally, or the DU being deployed centrally, with the CU being centralized at a higher level.

[0388] The AAU 12100 can implement transceiver functions and is called a transceiver unit 12100, corresponding to the transceiver module 1001 in Figure 10. Optionally, the transceiver unit 12100 can also be called a transceiver, a transceiver circuit, or a transceiver, and can include at least one antenna 12101 and a radio frequency unit 12102. Optionally, the transceiver unit 12100 can include a receiving unit and a transmitting unit. The receiving unit can correspond to a receiver (or receiver, receiving circuit), and the transmitting unit can correspond to a transmitter (or transmitter, transmitting circuit).

[0389] The CU and DU 12200 can implement internal processing functions and are referred to as processing units 12200, corresponding to processing module 1002 in Figure 10. Optionally, the processing unit 12200 can control network devices and can be referred to as a controller. The AAU can be physically located together with the CU and DU, or physically separated.

[0390] In addition, the network device is not limited to the form shown in Figure 12, and can also be other forms: for example, including a BBU and an adaptive radio unit (ARU), or including a BBU and an active antenna unit (AAU); it can also be customer premises equipment (CPE), or it can be other forms, which are not limited in this application.

[0391] In one example, the processing unit 12200 can be composed of one or more single boards, and multiple single boards can jointly support a wireless access network with a single access standard (such as an LTE network), or can respectively support wireless access networks with different access standards (such as an LTE network, a 5G network, a future network or other networks). The CU and DU 12200 also include a memory 12201 and a processor 12202. The memory 12201 is used to store necessary instructions and data. The processor 12202 is used to control the network device to perform necessary actions, such as controlling the network device to execute the operation process of the second device in the above method embodiment. The memory 12201 and the processor 12202 can serve one or more single boards. That is, a memory and a processor can be set separately on each single board. Alternatively, multiple single boards can share the same memory and processor. In addition, necessary circuits can also be set on each single board.

[0392] It should be understood that the network device 1200 shown in Figure 12 is capable of implementing the second device function involved in the method embodiments of Figures 2, 7 and 8. The operations and / or functions of the various units in the network device 1200 are respectively for implementing the corresponding processes performed by the network device in the method embodiments of the present application. To avoid repetition, detailed descriptions are appropriately omitted here. The structure of the network device illustrated in Figure 12 is only one possible form and should not constitute any limitation on the embodiments of the present application. The present application does not exclude the possibility of other forms of network device structures that may appear in the future.

[0393] The CU and DU 12200 can be used to perform the actions implemented within the second device described in the previous method embodiment, while the AAU 12100 can be used to perform the transceiver actions of the second device described in the previous method embodiment. For details, please refer to the description in the previous method embodiment and will not be repeated here.

[0394] The present application also provides a computer program product, which includes: computer program code, which, when running on a computer, enables the computer to execute the method of any one of the embodiments shown in Figures 2, 7 and 8.

[0395] The present application also provides a computer-readable medium storing a program code. When the program code runs on a computer, the computer executes the method of any one of the embodiments shown in FIG. 2 , FIG. 7 and FIG. 8 .

[0396] The present application also provides a communication system, which includes a first device and a second device. The first device is configured to execute some or all of the steps executed by the first device in the embodiments shown in Figures 2, 7, and 8, and the second device is configured to execute some or all of the steps executed by the second device in the embodiments shown in Figures 2, 7, and 8.

[0397] An embodiment of the present application further provides a chip device, including a processor, configured to call a computer program or computer instruction stored in the memory so that the processor executes the method of the embodiments shown in FIG. 2 , FIG. 7 , and FIG. 8 .

[0398] In one possible implementation, the input of the chip device corresponds to the receiving operation in the embodiments shown in FIG. 2 , FIG. 7 and FIG. 8 , and the output of the chip device corresponds to the sending operation in the embodiments shown in FIG. 2 , FIG. 7 and FIG. 8 .

[0399] Optionally, the processor is coupled to the memory via an interface.

[0400] Optionally, the chip device further includes a memory, in which computer programs or computer instructions are stored.

[0401] The processor mentioned in any of the above may be a general-purpose central processing unit, a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the method of the embodiments shown in Figures 2, 7, and 8. The memory mentioned in any of the above may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), etc.

[0402] Those skilled in the art can clearly understand that, for the sake of convenience and brevity of description, the explanation and beneficial effects of the relevant contents in any of the communication devices provided above can refer to the corresponding method embodiments provided above, and will not be repeated here.

[0403] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0404] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0405] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0406] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0407] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A communication method, characterized in that: The method comprises: The first device receives first information from the second device, where the first information is used to respectively instruct the first device whether to send local model parameters of the first model of the first device; The first device determines, according to the first information, some local model parameters of the first model to be sent, where the some local model parameters are obtained by training the first model; The first device sends the portion of local model parameters to the second device.

2. A communication method, characterized in that: The method comprises: The second device sends first information to the first device, where the first information is used to respectively instruct the first device whether to send local model parameters of the first model of the first device; The second device receives part of the local model parameters of the first model from the first device, where the part of the local model parameters is obtained by training the first model.

3. The method according to claim 1 or 2, characterized in that The local model parameters include local weight parameters of the first model.

4. The method according to claim 3, characterized in that The local weight parameters include local weights or local weight gradients of the first model.

5. The method according to any one of claims 1 to 4, characterized in that All local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

6. The method according to any one of claims 1 to 4, characterized in that All local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters.

7. The method according to claim 5 or 6, characterized in that The method further comprises: The first device receives N global model parameters of the first model or global model parameters of P layers of neurons of the first model from the second device.

8. The method according to claim 5 or 6, characterized in that The method further comprises: The second device sends N global model parameters of the first model or global model parameters of P layers of neurons in the first model to the first device.

9. The method according to claim 7 or 8, characterized in that The N global model parameters correspond one-to-one to the N local model parameters; wherein, the N first indication information and the N global model parameters are carried in the same signaling or different signalings. When the N first indication information and the N global model parameters are carried in the same signaling, the N global model parameters and the N first indication information are arranged at intervals, and the first indication information corresponding to the global model parameter is arranged adjacently after each global model parameter, or, the N global model parameters are arranged before the N first indication information.

10. The method according to claim 7 or 8, characterized in that The global model parameters of the P layer neurons correspond one-to-one to the local model parameters of the P layer neurons; wherein, the P second indication information and the global model parameters of the P layer neurons are carried in the same signaling or different signalings. When the P second indication information and the global model parameters of the P layer neurons are carried in the same signaling, the global model parameters of the P layer neurons and the P second indication information are arranged at intervals, and the second indication information corresponding to the global model parameters of each layer of neurons is arranged adjacent to the global model parameters of each layer of neurons, or, the global model parameters of the P layer neurons are arranged before the P second indication information.

11. The method according to any one of claims 1 to 4, characterized in that All local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; The first information includes a first identification bit and a layer sequence number of at least one first target layer in the P layer neurons, the first identification bit being used to indicate that the first device does not send local model parameters of the neurons of the at least one first target layer; or The first information includes a second identification bit and a layer sequence number of at least one second target layer in the P layer neurons, and the second identification bit is used to instruct the first device to send local model parameters of the neurons of the at least one second target layer.

12. A communication method, characterized in that: The method comprises: The first device determines part of local model parameters of the first model of the first device to be sent, where the part of local model parameters is obtained by training the first model; The first device sends the part of the local model parameters and first information to the second device, where the first information is used to instruct the first device to send the part of the local model parameters.

13. A communication method, characterized in that: The method comprises: The second device receives part of the local model parameters of the first model and first information from the first device, where the first information is used to instruct the first device to send the part of the local model parameters, where the part of the local model parameters is obtained by training the first model; The second device determines the part of local model parameters according to the first information.

14. The method according to claim 12 or 13, characterized in that The part of local model parameters includes local weight parameters of the first model.

15. The method according to claim 14, characterized in that The local weight parameters include local weights or local weight gradients of the first model.

16. The method according to any one of claims 12 to 15, characterized in that All local model parameters of the first model include N local model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N local model parameters. The first indication information corresponding to each local model parameter in the N local model parameters is used to indicate whether the first device sends the local model parameter.

17. The method according to any one of claims 12 to 15, characterized in that All local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information corresponds one-to-one to the local model parameters of the P layers of neurons. The second indication information corresponding to the local model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the first device sends the local model parameters of each layer of neurons.

18. The method according to any one of claims 12 to 15, characterized in that All local model parameters of the first model include local model parameters of P layers of neurons, where P is an integer greater than or equal to 1; The first information includes a first identification bit and a layer sequence number of at least one first target layer in the P-layer neurons, the first identification bit being used to instruct the first device not to send local model parameters of the neurons of the at least one first target layer; or, The first information includes a second identification bit and a layer sequence number of at least one second target layer in the P layer neurons, and the second identification bit is used to instruct the first device to send local model parameters of the neurons of the at least one second target layer.

19. The method according to any one of claims 12, 14 to 18, characterized in that The first device determines a portion of local model parameters of a first model of the first device to be sent, including: The first device determines the partial local model parameters based on at least one of the local model parameters obtained by the first device through the Rth round of training of the first model, the communication link status of the first device, and the computing power of the first device. The partial local model parameters are obtained by the first device through the R+1th round of training of the first model, where R is an integer greater than or equal to 1.

20. A communication method, characterized in that: The method comprises: A first device receives a portion of first global model parameters of a first model of the first device from a second device; The first device receives first information from the second device, where the first information is used to instruct the second device to send the portion of the first global model parameters; The first device updates the first model according to the first information and the part of the first global model parameters to obtain an updated first model.

21. A communication method, characterized in that: The method comprises: The second device sends a portion of first global model parameters of the first model of the first device to the first device; The second device sends first information to the first device, where the first information is used to instruct the second device to send the portion of the first global model parameters.

22. The method according to claim 20 or 21, characterized in that The portion of first global model parameters includes global weight parameters of the first model.

23. The method according to claim 22, characterized in that The global weight parameter includes the global weight or global weight gradient of the first model.

24. The method according to any one of claims 20 to 23, characterized in that All first global model parameters of the first model include N first global model parameters, where N is an integer greater than or equal to 2; the first information includes N first indication information, and the N first indication information corresponds one-to-one to the N first global model parameters. The first indication information corresponding to each first global model parameter in the N first global model parameters is used to indicate whether the second device sends the first global model parameter.

25. The method according to any one of claims 20 to 23, characterized in that All first global model parameters of the first model include the first global model parameters of P layers of neurons, where P is an integer greater than or equal to 1; the first information includes P second indication information, and the P second indication information correspond one-to-one to the first global model parameters of the P layers of neurons. The second indication information corresponding to the first global model parameters of each layer of neurons in the P layers of neurons is used to indicate whether the second device sends the first global model parameters of each layer of neurons.

26. The method according to any one of claims 20 to 23, characterized in that All first global model parameters of the first model include first global model parameters of P-layer neurons, where P is an integer greater than or equal to 1; The first information includes a first identification bit and a layer sequence number of at least one first target layer in the P layer neurons, the first identification bit being used to instruct the second device not to send the first global model parameter of the neurons of the at least one first target layer; or The first information includes a second identification bit and a layer sequence number of at least one second target layer in the P layer neurons, and the second identification bit is used to instruct the second device to send the first global model parameters of the neurons of the at least one second target layer.

27. The method according to any one of claims 20 to 26, characterized in that All first global model parameters of the first model include N first global model parameters obtained by the second device in the M+1 round of fusing local model parameters of multiple devices, where N is an integer greater than or equal to 2; the N first global model parameters correspond one-to-one to N second global model parameters, and the N second global model parameters are obtained by the second device in the M round of fusing local model parameters of multiple devices, where M is an integer greater than or equal to 1; among the part of the first global model parameters, the ratio of the change between each first global model parameter and the second global model parameter corresponding to the first global model parameter to the second global model parameter is greater than the first ratio.

28. A first device, characterized in that: The first device includes a transceiver module and a processing module; The transceiver module is used to perform the transceiver operation according to any one of claims 1, 3 to 7, and 9 to 11, and the processing module is used to perform the processing operation according to any one of claims 1, 3 to 7, and 9 to 11; or The transceiver module is used to perform the transceiver operation according to any one of claims 12, 14 to 19, and the processing module is used to perform the processing operation according to any one of claims 12, 14 to 19; or The transceiver module is used to perform the transceiver operation according to any one of claims 20, 22 to 27, and the processing module is used to perform the processing operation according to any one of claims 20, 22 to 27.

29. A second device, characterized in that: The second device includes a transceiver module; The transceiver module is configured to perform the transceiver operation according to any one of claims 2 to 6 and 8 to 11; or The transceiver module is configured to perform the transceiver operation as claimed in any one of claims 21 to 27.

30. A second device, characterized in that: The second device includes a transceiver operation and a processing module, the transceiver module is used to perform the transceiver operation according to any one of claims 13 to 18, and the processing module is used to perform the processing operation according to any one of claims 13 to 18.

31. A device, characterized in that The device includes a processor; the processor is used to execute a computer program or computer instructions in a memory to perform the method according to any one of claims 1, 3 to 7, 9 to 11; or, the processor is used to execute a computer program or computer instructions in the memory to perform the method according to any one of claims 12, 14 to 19; or, the processor is used to execute a computer program or computer instructions in the memory to perform the method according to claims 20, 22 to 27; or, the processor is used to execute a computer program or computer instructions in the memory to perform the method according to claims 2 to 6, 8 to 11; or, the processor is used to execute a computer program or computer instructions in the memory to perform the method according to claims 13 to 18; or, the processor is used to execute a computer program or computer instructions in the memory to perform the method according to claims 21 to 27.

32. The device according to claim 31, characterized in that The apparatus further comprises the memory.

33. A computer-readable storage medium, characterized in that A computer program is stored thereon, and when the computer program is executed by the device, the device performs the method as claimed in any one of claims 1 to 11, or the method as claimed in any one of claims 12 to 19, or the method as claimed in any one of claims 20 to 27.

34. A computer program product, characterized in that When the computer program product is run on a computer, the computer is enabled to perform the method according to any one of claims 1 to 11, or the computer is enabled to perform the method according to any one of claims 12 to 19, or the computer is enabled to perform the method according to any one of claims 20 to 27.

Images