While neural networks have acted as a strong unifying force in the design of modern AI systems, the neural network architectures themselves remain highly heterogeneous due to the variety of tasks to be solved. In this chapter, we explore how to adapt the Layer-wise Relevance Propagation (LRP) technique used for explaining the predictions of feed-forward networks to the LSTM architecture used for sequential data modeling and forecasting. The special accumulators and gated interactions present in the LSTM require both a new propagation scheme and an extension of the underlying theoretical framework to deliver faithful explanations.
@article{arxiv.1909.12114,
title = {Explaining and Interpreting LSTMs},
author = {Leila Arras and Jose A. Arjona-Medina and Michael Widrich and Grégoire Montavon and Michael Gillhofer and Klaus-Robert Müller and Sepp Hochreiter and Wojciech Samek},
journal= {arXiv preprint arXiv:1909.12114},
year = {2019}
}
Comments
28 pages, 7 figures, book chapter, In: Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, LNCS volume 11700, Springer 2019. arXiv admin note: text overlap with arXiv:1806.07857