MEME: Generating RNN Model Explanations via Model Extraction
Abstract
Recurrent Neural Networks (RNNs) have achieved remarkable performance on a range of tasks. A key step to further empowering RNN-based approaches is improving their explainability and interpretability. In this work we present MEME: a model extraction approach capable of approximating RNNs with interpretable models represented by human-understandable concepts and their interactions. We demonstrate how MEME can be applied to two multivariate, continuous data case studies: Room Occupation Prediction, and In-Hospital Mortality Prediction. Using these case-studies, we show how our extracted models can be used to interpret RNNs both locally and globally, by approximating RNN decision-making via interpretable concept interactions.
Cite
@article{arxiv.2012.06954,
title = {MEME: Generating RNN Model Explanations via Model Extraction},
author = {Dmitry Kazhdan and Botty Dimanov and Mateja Jamnik and Pietro Liò},
journal= {arXiv preprint arXiv:2012.06954},
year = {2021}
}
Comments
Presented at the HAMLETS workshop at the 34th Conference on Neural Information Processing Systems (NeurIPS 2020)