Greedy Optimization Provably Wins the Lottery: Logarithmic Number of Winning Tickets is Enough
From MaRDI portal
Publication:6352609
arXiv2010.15969MaRDI QIDQ6352609
Author name not available (Why is that?)
Publication date: 29 October 2020
Abstract: Despite the great success of deep learning, recent works show that large deep neural networks are often highly redundant and can be significantly reduced in size. However, the theoretical question of how much we can prune a neural network given a specified tolerance of accuracy drop is still open. This paper provides one answer to this question by proposing a greedy optimization based pruning method. The proposed method has the guarantee that the discrepancy between the pruned network and the original network decays with exponentially fast rate w.r.t. the size of the pruned network, under weak assumptions that apply for most practical settings. Empirically, our method improves prior arts on pruning various network architectures including ResNet, MobilenetV2/V3 on ImageNet.
Has companion code repository: https://github.com/lushleaf/Network-Pruning-Greedy-Forward-Selection
This page was built for publication: Greedy Optimization Provably Wins the Lottery: Logarithmic Number of Winning Tickets is Enough
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6352609)
