Mixed Precision Training of Convolutional Neural Networks using Integer Operations

TL;DR

This paper demonstrates that training state-of-the-art CNNs on ImageNet-1K using 16-bit integer operations on general purpose hardware can match or surpass FP32 accuracy with significantly improved training throughput.

Contribution

It introduces a novel integer training scheme with shared exponent representation and efficient convolution kernels, enabling SOTA CNN training with 16-bit integers on GP hardware.

Findings

Achieved or exceeded SOTA accuracy with INT16 training on ImageNet-1K.

Realized a 1.8X increase in training throughput over FP32.

First to report INT16 training results for SOTA CNNs on GP hardware.

Abstract

The state-of-the-art (SOTA) for mixed precision training is dominated by variants of low precision floating point operations, and in particular, FP16 accumulating into FP32 Micikevicius et al. (2017). On the other hand, while a lot of research has also happened in the domain of low and mixed-precision Integer training, these works either present results for non-SOTA networks (for instance only AlexNet for ImageNet-1K), or relatively small datasets (like CIFAR-10). In this work, we train state-of-the-art visual understanding neural networks on the ImageNet-1K dataset, with Integer operations on General Purpose (GP) hardware. In particular, we focus on Integer Fused-Multiply-and-Accumulate (FMA) operations which take two pairs of INT16 operands and accumulate results into an INT32 output.We propose a shared exponent representation of tensors and develop a Dynamic Fixed Point (DFP) scheme suitable for common neural network operations. The nuances of developing an efficient integer convolution kernel is examined, including methods to handle overflow of the INT32 accumulator. We implement CNN training for ResNet-50, GoogLeNet-v1, VGG-16 and AlexNet; and these networks achieve or exceed SOTA accuracy within the same number of iterations as their FP32 counterparts without any change in hyper-parameters and with a 1.8X improvement in end-to-end training throughput. To the best of our knowledge these results represent the first INT16 training results on GP hardware for ImageNet-1K dataset using SOTA CNNs and achieve highest reported accuracy using half-precision