Bayesian Sparsification of Gated Recurrent Neural Networks
arXiv:1812.05692
Abstract
Bayesian methods have been successfully applied to sparsify weights of neural networks and to remove structure units from the networks, e. g. neurons. We apply and further develop this approach for gated recurrent architectures. Specifically, in addition to sparsification of individual weights and neurons, we propose to sparsify preactivations of gates and information flow in LSTM. It makes some gates and information flow components constant, speeds up forward pass and improves compression. Moreover, the resulting structure of gate sparsity is interpretable and depends on the task. Code is available on github: https://github.com/tipt0p/SparseBayesianRNN
arXiv commentPublished in Workshop on Compact Deep Neural Networks with industrial applications, NeurIPS 2018
References in corpus (6)
- Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
- A Simple Way to Initialize Recurrent Networks of Rectified Linear Units
- Bayesian Compression for Deep Learning
- Exploring Sparsity in Recurrent Neural Networks
- Learning Intrinsic Sparse Structures within Long Short-Term Memory
- Compressing Recurrent Neural Network with Tensor Train