Increasing the Interpretability of Recurrent Neural Networks Using Hidden Markov Models
Abstract
As deep neural networks continue to revolutionize various application domains, there is increasing interest in making these powerful models more understandable and interpretable, and narrowing down the causes of good and bad predictions. We focus on recurrent neural networks (RNNs), state of the art models in speech recognition and translation. Our approach to increasing interpretability is by combining an RNN with a hidden Markov model (HMM), a simpler and more transparent model. We explore various combinations of RNNs and HMMs: an HMM trained on LSTM states; a hybrid model where an HMM is trained first, then a small LSTM is given HMM state distributions and trained to fill in gaps in the HMM's performance; and a jointly trained hybrid model. We find that the LSTM and HMM learn complementary information about the features in the text.
Cite
@article{arxiv.1606.05320,
title = {Increasing the Interpretability of Recurrent Neural Networks Using Hidden Markov Models},
author = {Viktoriya Krakovna and Finale Doshi-Velez},
journal= {arXiv preprint arXiv:1606.05320},
year = {2016}
}
Comments
presented at 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016), New York, NY