Open Peer Review. Open Publishing. Open Access. Open Discussion. Open Directory. Open Recommendations. Open API. Open Source.
Noise-Based Regularizers for Recurrent Neural Networks
Adji B. Dieng, Jaan Altosaar, Rajesh Ranganath, David M. Blei
Feb 15, 2018 (modified: Feb 15, 2018)ICLR 2018 Conference Blind Submissionreaders: everyoneShow Bibtex
Abstract:Recurrent neural networks (RNNs) are powerful models for sequential data. They can approximate arbitrary computations, and have been used successfully in domains such as text and speech. However, the flexibility of RNNs makes them susceptible to overfitting and regularization is important. We develop a noise-based regularization method for RNNs. The idea is simple and easy to implement: we inject noise in the hidden units of the RNN and then maximize the original RNN's likelihood averaged over the injected noise. On a language modeling benchmark, our method achieves better performance than the deterministic RNN and the variational dropout.
Enter your feedback below and we'll get back to you as soon as possible.