HIGH-LATENCY NETWORK ENVIRONMENT ROBUST FEDERATED LEARNING TRAINING METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM

摘要

A high-latency network environment robust federated learning training method and apparatus, a computer device, and a storage medium, relating to artificial intelligence technology, and comprising: obtaining a current system time, and if encrypted data uploaded by several data upload terminals is not received, obtaining corresponding target data upload terminals (S110, S120); obtaining current network latency values of the target data upload terminals, so as to obtain a maximum network latency value (S130); calculating a latency stride according to the maximum network latency value and a unit timing interval stride (S140); adding the current system time and the latency stride to obtain a target system time, and if the current time is the target system time and target encrypted data uploaded by the target data upload terminals is not received, stopping local federated learning training, until the target encrypted data uploaded by all the target data upload terminals is received, then resuming the local federated learning training (S150, S160). The present invention maintains the training efficiency of federated learning via a latency sparse update means in the case of network latency.