期刊
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
卷 31, 期 10, 页码 3828-3838出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TNNLS.2019.2946636
关键词
Sparse matrices; Matrices; Indexes; Neural networks; Task analysis; Bayes methods; Learning systems; Deep neural networks (DNNs); memory efficiency; network compression; sparse representation
类别
资金
- NSFC [61876095, 61751308]
- Beijing Academy of Artificial Intelligence (BAAI)
Modern deep neural networks (DNNs) are usually overparameterized and composed of a large number of learnable parameters. One of a few effective solutions attempts to compress DNN models via learning sparse weights and connections. In this article, we follow this line of research and present an alternative framework of learning sparse DNNs, with the assistance of matrix factorization. We provide an underlying principle for substituting the original parameter matrices with the multiplications of highly sparse ones, which constitutes the theoretical basis of our method. Experimental results demonstrate that our method substantially outperforms previous states of the arts for compressing various DNNs, giving rich empirical evidence in support of its effectiveness. It is also worth mentioning that, unlike many other works that focus on feedforward networks like multi-layer perceptrons and convolutional neural networks only, we also evaluate our method on a series of recurrent networks in practice.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据