A Systematic DNN Weight Pruning Framework using Alternating Direction Method of Multipliers

TL;DR

This paper introduces a systematic DNN weight pruning framework using ADMM, providing guarantees on sparsity, convergence, and outperforming prior heuristic methods in reducing model size and computation without accuracy loss.

Contribution

The paper formulates DNN weight pruning as a nonconvex optimization problem and applies ADMM for systematic, guaranteed, and efficient pruning, outperforming prior heuristic approaches.

Findings

Achieves 71.2x weight reduction on LeNet-5 without accuracy loss.

Achieves 21x weight reduction on AlexNet without accuracy loss.

Reduces total computation by five times in convolutional layers.

Abstract

Weight pruning methods for deep neural networks (DNNs) have been investigated recently, but prior work in this area is mainly heuristic, iterative pruning, thereby lacking guarantees on the weight reduction ratio and convergence time. To mitigate these limitations, we present a systematic weight pruning framework of DNNs using the alternating direction method of multipliers (ADMM). We first formulate the weight pruning problem of DNNs as a nonconvex optimization problem with combinatorial constraints specifying the sparsity requirements, and then adopt the ADMM framework for systematic weight pruning. By using ADMM, the original nonconvex optimization problem is decomposed into two subproblems that are solved iteratively. One of these subproblems can be solved using stochastic gradient descent, the other can be solved analytically. Besides, our method achieves a fast convergence rate. The weight pruning results are very promising and consistently outperform the prior work. On the LeNet-5 model for the MNIST data set, we achieve 71.2 times weight reduction without accuracy loss. On the AlexNet model for the ImageNet data set, we achieve 21 times weight reduction without accuracy loss. When we focus on the convolutional layer pruning for computation reductions, we can reduce the total computation by five times compared with the prior work (achieving a total of 13.4 times weight reduction in convolutional layers). Our models and codes are released at https://github.com/KaiqiZhang/admm-pruning