Stochastic Weight Averaging in Parallel: Large-Batch Training That Generalizes WellDownload PDF

Sep 25, 2019 (edited Mar 11, 2020)ICLR 2020 Conference Blind SubmissionReaders: Everyone
  • Original Pdf: pdf
  • Keywords: Large batch training, Distributed neural network training, Stochastic Weight Averaging
  • TL;DR: We propose SWAP, a distributed algorithm for large-batch training of neural networks.
  • Abstract: We propose Stochastic Weight Averaging in Parallel (SWAP), an algorithm to accelerate DNN training. Our algorithm uses large mini-batches to compute an approximate solution quickly and then refines it by averaging the weights of multiple models computed independently and in parallel. The resulting models generalize equally well as those trained with small mini-batches but are produced in a substantially shorter time. We demonstrate the reduction in training time and the good generalization performance of the resulting models on the computer vision datasets CIFAR10, CIFAR100, and ImageNet.
9 Replies