Evaluating Attribution Methods using White-Box LSTMs
Proceedings of the 2020 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP
Venue: BlackboxNLP
Type: Workshop
Architectures
Formal Languages
Interpretability
Author
Affiliation
Sophie Hao
Yale University
Published
November 20, 2020
Abstract
Interpretability methods for neural networks are difficult to evaluate because we do not understand the black-box models typically used to test them. This paper proposes a framework in which interpretability methods are evaluated using manually constructed networks, which we call white-box networks, whose behavior is understood a priori. We evaluate five methods for producing attribution heatmaps by applying them to white-box LSTM classifiers for tasks based on formal languages. Although our white-box classifiers solve their tasks perfectly and transparently, we find that all five attribution methods fail to produce the expected model explanations.