To address the large model size and intensive computation requirement of deep neural networks (DNNs), weight pruning techniques have been proposed and generally fall into two categories, i.e., static regularization-based pruning and dynamic regularization-based pruning. However, the former method currently suffers either complex workloads or accuracy degradation, while the latter one takes a long time to tune the parameters to achieve the desired pruning rate without accuracy loss. In this paper, we propose a unified DNN weight pruning framework with dynamically updated regularization terms bounded by the designated constraint, which can generate both non-structured sparsity and different kinds of structured sparsity. We also extend our method to an integrated framework for the combination of different DNN compression tasks.
@article{arxiv.2004.05531,
title = {A Unified DNN Weight Compression Framework Using Reweighted Optimization Methods},
author = {Tianyun Zhang and Xiaolong Ma and Zheng Zhan and Shanglin Zhou and Minghai Qin and Fei Sun and Yen-Kuang Chen and Caiwen Ding and Makan Fardad and Yanzhi Wang},
journal= {arXiv preprint arXiv:2004.05531},
year = {2020}
}