SCOP: Scientific Control for Reliable Neural Network Pruning

TL;DR

This paper introduces a scientifically controlled neural network pruning method using knockoff features as control groups, leading to more reliable filter importance assessment and significant model compression with minimal accuracy loss.

Contribution

The paper proposes a novel pruning algorithm that incorporates scientific control via knockoff features, improving the reliability of filter importance evaluation in neural networks.

Findings

Reduces 57.8% parameters in ResNet-101 on ImageNet.

Reduces 60.2% FLOPs with only 0.01% accuracy loss.

Outperforms state-of-the-art pruning methods.

Abstract

This paper proposes a reliable neural network pruning algorithm by setting up a scientific control. Existing pruning methods have developed various hypotheses to approximate the importance of filters to the network and then execute filter pruning accordingly. To increase the reliability of the results, we prefer to have a more rigorous research design by including a scientific control group as an essential part to minimize the effect of all factors except the association between the filter and expected network output. Acting as a control group, knockoff feature is generated to mimic the feature map produced by the network filter, but they are conditionally independent of the example label given the real feature map. We theoretically suggest that the knockoff condition can be approximately preserved given the information propagation of network layers. Besides the real feature map on an intermediate layer, the corresponding knockoff feature is brought in as another auxiliary input signal for the subsequent layers. Redundant filters can be discovered in the adversarial process of different features. Through experiments, we demonstrate the superiority of the proposed algorithm over state-of-the-art methods. For example, our method can reduce 57.8% parameters and 60.2% FLOPs of ResNet-101 with only 0.01% top-1 accuracy loss on ImageNet. The code is available at https://github.com/huawei-noah/Pruning/tree/master/SCOP_NeurIPS2020.