We propose a novel network pruning approach by information preserving of pre-trained network weights (filters). Network pruning with the information preserving is formulated as a matrix sketch problem, which is efficiently solved by the off-the-shelf Frequent Direction method. Our approach, referred to as FilterSketch, encodes the second-order information of pre-trained weights, which enables the representation capacity of pruned networks to be recovered with a simple fine-tuning procedure. FilterSketch requires neither training from scratch nor data-driven iterative optimization, leading to a several-orders-of-magnitude reduction of time cost in the optimization of pruning. Experiments on CIFAR-10 show that FilterSketch reduces 63.3% of FLOPs and prunes 59.9% of network parameters with negligible accuracy cost for ResNet-110. On ILSVRC-2012, it reduces 45.5% of FLOPs and removes 43.0% of parameters with only 0.69% accuracy drop for ResNet-50. Our code and pruned models can be found at https://github.com/lmbxmu/FilterSketch.
@article{arxiv.2001.08514,
title = {Filter Sketch for Network Pruning},
author = {Mingbao Lin and Liujuan Cao and Shaojie Li and Qixiang Ye and Yonghong Tian and Jianzhuang Liu and Qi Tian and Rongrong Ji},
journal= {arXiv preprint arXiv:2001.08514},
year = {2021}
}
Comments
Accepted by IEEE Transactions on Neural Networks and Learning Systems (IEEE TNNLS)