NeurIPS 2022

Timezone: »

Interactive Rationale Extraction for Text Classification

Jiayi Dai · Mi-Young Kim · Randolph Goebel

in Trustworthy and Socially Responsible Machine Learning »

Event URL: https://openreview.net/forum?id=zaJsDuwwdlJ »

Deep neural networks show superior performance in text classification tasks, but their poor interpretability and explainability can cause trust issues. For text classification problems, the identification of textual sub-phrases or ``rationales'' is one strategy for attempting to find the most influential portions of text, which can be conveyed as critical in making classification decisions. Selective models for rationale extraction faithfully explain a neural classifier's predictions by training a rationale generator and a text classifier jointly: the generator identifies rationales and the classifier predicts a category solely based on the rationales. The selected rationales are then viewed as the explanations for the classifier's predictions. Through exchange of such explanations, humans interact to achieve higher performances in problem solving. To imitate the interactive process of humans, we propose a simple interactive rationale extraction architecture that selects a pair of rationales and then makes predictions from two independently trained selective models. We show how this architecture outperforms both base models for text classification tasks on datasets IMDB movie reviews and 20 Newsgroups in terms of predictive performance.

[ Poster] [ OpenReview] [ Topia]

Author Information

Jiayi Dai (University of Alberta)

Mi-Young Kim (University of Alberta)

Randolph Goebel (University of Alberta, Alberta Machine Intelligence Institute)

R.G. (Randy) Goebel is currently professor and chair in the Department of Computing Science at the University of Alberta He received the B.Sc. (Computer Science), M.Sc. (Computing Science), and Ph.D. (Computer Science) from the Universities of Regina, Alberta, and British Columbia, respectively. Professor Goebel's research is focused on the theory and application of intelligent systems. His theoretical work on abduction, hypothetical reasoning and belief revision is internationally well know, and his recent application of practical belief revision and constraint programming to scheduling, layout, and web mining is now having industrial impact. He is one of the founders of the Alberta Ingenuity Centre for Machine Learning (AICML), and is now working on applications of machine learning to various problems, including web visualization and scheduling. Randy has previously held faculty appointments at the University of Waterloo and the University of Tokyo, and is actively involved in academic and industrial collaborative research projects in Canada, Australia, Malaysia, Europe and Japan.

Author Information

Jiayi Dai (University of Alberta)

Mi-Young Kim (University of Alberta)

Randolph Goebel (University of Alberta, Alberta Machine Intelligence Institute)

More from the Same Authors