Model training method and device, electronic equipment, storage medium and program product

Abstract

The invention provides a model training method and device, electronic equipment, a storage medium and a program product, and relates to the technical field of artificial intelligence, in particular to the technical field of distributed computing. The method comprises the following steps: for a round of training of a target global model in a plurality of global models, selecting at least two target terminal devices from a plurality of terminal devices according to the time required for training each global model; sending global model parameters of the target global model to at least two target terminal devices; and receiving local model parameters sent by the at least two target terminal devices, and updating the target global model parameters according to the local model parameters sent by the at least two target terminal devices, the local model parameters being obtained by training the target global model by the at least two target terminal devices according to the local training samples. The training efficiency of the plurality of global models is improved.

Description

technical field [0001] The present disclosure relates to distributed computing technology in the technical field of artificial intelligence, and in particular to a model training method, device, electronic equipment, storage medium and program product. Background technique [0002] Federated learning is a new distributed learning mechanism that uses distributed data and computing resources for collaborative training of machine learning models. A federated learning system usually includes a server and multiple terminal devices. In federated learning, the server sends the global model to be trained to each terminal device, and each terminal device uses local private data to train and update model parameters, and the updated The model parameters are uploaded to the server, and finally the server aggregates the updated model parameters of each terminal device to obtain a new global model, and repeats the above training for multiple rounds until the global model reaches convergen...

AI Technical Summary

Problems solved by technology

[0003] In multi-task federated learning, there are multiple global models to be trained in the federated learning system. If each terminal device can only train one global model at the same time, this will undoubtedly increase the waiting time of other global models and the training efficiency is extremely low. To this end, you can choose to have multiple global models trained in parallel across multiple end devices

Images