On the Reduction of Computational Complexity of Deep Convolutional Neural Networks
Open Access
Abstract
Deep convolutional neural networks (ConvNets), which are at the heart of many new emerging applications, achieve remarkable performance in audio and visual recognition tasks. Unfortunately, achieving accuracy often implies significant computational costs, limiting deployability. In modern ConvNets it is typical for the convolution layers to consume the vast majority of computational resources during inference. This has made the acceleration of these layers an important research area in academia and industry. In this paper, we examine the effects of co-optimizing the internal structures of the convolutional layers and underlying implementation of fundamental convolution operation. We demonstrate that a combination of these methods can have a big impact on the overall speedup of a ConvNet, achieving a ten-fold increase over baseline. We also introduce a new class of fast one-dimensional (1D) convolutions for ConvNets using the Toom–Cook algorithm. We show that our proposed scheme is mathematically well-grounded, robust, and does not require any time-consuming retraining, while still achieving speedups solely from convolutional layers with no loss in baseline accuracy.Keywords
Funding Information
- Engineering and Physical Sciences Research Council
This publication has 20 references indexed in Scilit:
- Spectral shape classification: A deep learning approachJournal of Visual Communication and Image Representation, 2017
- Eyeriss: A Spatial Architecture for Energy-Efficient Dataflow for Convolutional Neural Networks2016
- Deep Residual Learning for Image Recognition2016
- Fast Algorithms for Convolutional Neural Networks2016
- Fast ConvNets Using Group-Wise Brain Damage2016
- You Only Look Once: Unified, Real-Time Object Detection2016
- Convolutional Neural Networks for Speech RecognitionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Deep Learning Face Representation from Predicting 10,000 Classes2014
- Learning Separable Filters2013
- The Approximation of One Matrix by Another of Lower RankPsychometrika, 1936