Shift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions
This addresses efficiency bottlenecks in neural networks for computer vision tasks, offering a novel method to reduce model size and computation while maintaining or improving performance.
The paper tackles the high computational and parameter costs of spatial convolutions in neural networks by introducing a parameter-free, FLOP-free 'shift' operation as an alternative, achieving improved accuracy on CIFAR10 and CIFAR100 with 60% fewer parameters and demonstrating resilience on ImageNet.
Neural networks rely on convolutions to aggregate spatial information. However, spatial convolutions are expensive in terms of model size and computation, both of which grow quadratically with respect to kernel size. In this paper, we present a parameter-free, FLOP-free "shift" operation as an alternative to spatial convolutions. We fuse shifts and point-wise convolutions to construct end-to-end trainable shift-based modules, with a hyperparameter characterizing the tradeoff between accuracy and efficiency. To demonstrate the operation's efficacy, we replace ResNet's 3x3 convolutions with shift-based modules for improved CIFAR10 and CIFAR100 accuracy using 60% fewer parameters; we additionally demonstrate the operation's resilience to parameter reduction on ImageNet, outperforming ResNet family members. We finally show the shift operation's applicability across domains, achieving strong performance with fewer parameters on classification, face verification and style transfer.