I am a PhD student in the UNC-NLP group at UNC Chapel Hill. - peterbhase.

This is the repository for the paper here, including code for measuring OOD-ness of explanation counterfactuals and several new search methods for ...

Code for "Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?" - peterbhase/InterpretableNLP-ACL2020.

Code for paper "When Can Models Learn From Explanations? A Formal Framework for Understanding the Roles of Explanation Data" - peterbhase/ExplanationRoles.

Code for paper "Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?" - peterbhase/LAS-NL-

To view and interpret model prototypes, which can be done without training a model using our provided images, use view_protos.ipynb and nearest_neighbors.ipynb ...

The code is compatible with Python 3.6+. data_analysis.Rmd is an R markdown file that makes all the plots in the paper. The required packages can be installed ...

Full-stack web dev working on GitHub Docs. Formerly Mozilla, MDN. Loves Python, PWAs, Postgres, Node, Search, and wants to learn everything. - peterbe.