Methods for interpreting and understanding deep neural networks

作者：Grégoire Montavon, Wojciech Samek, Klaus‐Robert Müller · 发表于：Digital Signal Processing · 年份：2017 · DOI：10.1016/j.dsp.2017.10.011 · 被引用次数：2771 · 研究领域：Explainable Artificial Intelligence (XAI)、Neural Networks and Applications、Adversarial Robustness in Machine Learning

This paper provides an entry point to the problem of interpreting a deep neural network model and explaining its predictions. It is based on a tutorial given at ICASSP 2017. As a tutorial paper, the set of methods covered here is not exhaustive, but sufficiently representative to discuss a number of questions in interpretability, technical challenges, and possible applications. The second part of the tutorial focuses on the recently proposed layer-wise relevance propagation (LRP) technique, for which we provide theory, recommendations, and tricks, to make most efficient use of it on real data.