Fast Convolution Algorithm for Convolutional Neural Networks

Recent advances in computing power made possible by developments of faster general-purpose graphics processing units (GPGPUs) have increased the complexity of convolutional neural network (CNN) models. However, because of the limited applications of the existing GPGPUs, CNN accelerators are becoming more important. The current accelerators focus on improvement in memory scheduling and architectures. Thus, the number of multiplier-accumulator (MAC) operations is not reduced. In this study, a new convolution layer operation algorithm is proposed using the coarse-to-fine method instead of hardware or architecture approaches. This algorithm is shown to reduce the MAC operations by 33%. However, the accuracy of the Top 1 is decreased only by 3% and the Top 5 only by 1%.

I. Introduction

Computer vision technology using deep learning has recently been utilized in such areas such medicine, entertainment, robots, and vehicles. These deep-learning technologies mimic the neural network within the brain. Among the deep neural networks, the convolutional neural network (CNN), in particular, has shown promising performance in such areas as object recognition and image classification [1] .