# Do Human Rationales Improve Machine Explanations?

## TL;DR

This paper demonstrates that incorporating human rationales in training improves the quality of machine explanations, making them more understandable to humans, especially in CNN-based text classification.

## Contribution

It connects learning with rationales to explainable AI, showing that human-provided explanations enhance machine explanation quality evaluated by humans.

## Key findings

- Supervised attention yields better explanations than unsupervised attention.
- Learning with rationales improves explanation quality as judged by humans.
- The approach enhances interpretability in CNN-based text classification.

## Abstract

Work on "learning with rationales" shows that humans providing explanations to a machine learning system can improve the system's predictive accuracy. However, this work has not been connected to work in "explainable AI" which concerns machines explaining their reasoning to humans. In this work, we show that learning with rationales can also improve the quality of the machine's explanations as evaluated by human judges. Specifically, we present experiments showing that, for CNN- based text classification, explanations generated using "supervised attention" are judged superior to explanations generated using normal unsupervised attention.