PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions

TL;DR

This paper introduces PerforatedCNNs, a method that reduces CNN computational costs by selectively skipping evaluations in convolutional layers, enabling faster inference on low-power devices.

Contribution

It presents a novel perforation technique inspired by source code optimization, significantly accelerating CNNs while maintaining accuracy.

Findings

Achieves 2x-4x speedup on AlexNet and VGG-16

Perforation complements existing acceleration methods

Reduces computational cost without substantial accuracy loss

Abstract

We propose a novel approach to reduce the computational cost of evaluation of convolutional neural networks, a factor that has hindered their deployment in low-power devices such as mobile phones. Inspired by the loop perforation technique from source code optimization, we speed up the bottleneck convolutional layers by skipping their evaluation in some of the spatial positions. We propose and analyze several strategies of choosing these positions. We demonstrate that perforation can accelerate modern convolutional networks such as AlexNet and VGG-16 by a factor of 2x - 4x. Additionally, we show that perforation is complementary to the recently proposed acceleration method of Zhang et al.