EAPruning: Evolutionary Pruning for Vision Transformers and CNNs

TL;DR

This paper introduces EAPruning, a simple evolutionary pruning method applicable to both vision transformers and CNNs, achieving significant FLOPs reduction and speedup without extensive domain expertise or hyperparameter tuning.

Contribution

We propose a universal evolutionary pruning approach that simplifies structured pruning for diverse neural network architectures, enabling efficient deployment.

Findings

50% FLOPS reduction for ResNet50 and MobileNetV1

Nearly 40% FLOPs reduction for DeiT-Base

Achieved 1.37x and 1.34x speedup for CNNs, 1.4x for transformers

Abstract

Structured pruning greatly eases the deployment of large neural networks in resource-constrained environments. However, current methods either involve strong domain expertise, require extra hyperparameter tuning, or are restricted only to a specific type of network, which prevents pervasive industrial applications. In this paper, we undertake a simple and effective approach that can be easily applied to both vision transformers and convolutional neural networks. Specifically, we consider pruning as an evolution process of sub-network structures that inherit weights through reconstruction techniques. We achieve a 50% FLOPS reduction for ResNet50 and MobileNetV1, leading to 1.37x and 1.34x speedup respectively. For DeiT-Base, we reach nearly 40% FLOPs reduction and 1.4x speedup. Our code will be made available.