Speculative Backpropagation for CNN Parallel Training

Sangwoo Park, Taeweon Suh

Research output: Contribution to journalArticlepeer-review

Abstract

The parallel learning in neural networks can greatly shorten the training time. Its prior efforts were mostly limited to distributing inputs to multiple computing engines. It is because the gradient descent algorithm in the neural network training is inherently sequential. This paper proposes a novel CNN parallel training method for image recognition. It overcomes the sequential property of the gradient descent and enables the parallel training with the speculative backpropagation. We found that the Softmax and ReLU outcomes in the forward propagation for the same labels are likely to be very similar. This characteristic makes it possible to perform the forward and backward propagation simultaneously. We implemented the proposed parallel model with CNNs in both software and hardware, and evaluated its performance. The parallel training reduces the training time by 34% in CIFAR-100 without the loss of the prediction accuracy compared to the sequential training. In many cases, it even improves the accuracy.

Original languageEnglish
Article number9272337
Pages (from-to)215365-215374
Number of pages10
JournalIEEE Access
Volume8
DOIs
Publication statusPublished - 2020

Keywords

  • Deep learning
  • FPGA
  • parallel training
  • speculative backpropagation
  • training accelerator

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Fingerprint Dive into the research topics of 'Speculative Backpropagation for CNN Parallel Training'. Together they form a unique fingerprint.

Cite this