Learning Explanations from Language Data

08/13/2018

∙

PatternAttribution is a recent method, introduced in the vision domain, that explains classifications of deep neural networks. We demonstrate that it also generates meaningful interpretations in the language domain.

READ FULL TEXT

Learning Explanations from Language Data

Decision problem of some bundled FOML fragments

Explaining the Road Not Taken

Explaining Deep Neural Networks

Memory networks for consumer protection:unfairness exposed

Interpreting Vision and Language Generative Models with Semantic Visual Priors

In-Distribution Interpretability for Challenging Modalities

Leveraging Sparse Linear Layers for Debuggable Deep Networks

Learning Explanations from Language Data

Related Research

Decision problem of some bundled FOML fragments

Explaining the Road Not Taken

Explaining Deep Neural Networks

Memory networks for consumer protection:unfairness exposed

Interpreting Vision and Language Generative Models with Semantic Visual Priors

In-Distribution Interpretability for Challenging Modalities

Leveraging Sparse Linear Layers for Debuggable Deep Networks