Rationalization: A Neural Machine Translation Approach to Generating Natural Language Explanations

Upol Ehsan, Brent Harrison, Larry Chan, Mark O. Riedl

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

72 Scopus citations

Abstract

We introduce AI rationalization, an approach for generating explanations of autonomous system behavior as if a human had performed the behavior. We describe a rationalization technique that uses neural machine translation to translate internal state-action representations of an autonomous agent into natural language. We evaluate our technique in the Frogger game environment, training an autonomous game playing agent to rationalize its action choices using natural language. A natural language training corpus is collected from human players thinking out loud as they play the game. We motivate the use of rationalization as an approach to explanation generation and show the results of two experiments evaluating the effectiveness of rationalization. Results of these evaluations show that neural machine translation is able to accurately generate rationalizations that describe agent behavior, and that rationalizations are more satisfying to humans than other alternative methods of explanation.

Original languageEnglish
Title of host publicationAIES 2018 - Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society
Pages81-87
Number of pages7
ISBN (Electronic)9781450360128
DOIs
StatePublished - Dec 27 2018
Event1st AAAI/ACM Conference on AI, Ethics, and Society, AIES 2018 - New Orleans, United States
Duration: Feb 2 2018Feb 3 2018

Publication series

NameAIES 2018 - Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society

Conference

Conference1st AAAI/ACM Conference on AI, Ethics, and Society, AIES 2018
Country/TerritoryUnited States
CityNew Orleans
Period2/2/182/3/18

Bibliographical note

Publisher Copyright:
© 2018 ACM.

Keywords

  • ai rationalization
  • artificial intelligence
  • explainable ai
  • interpretability
  • machine learning
  • transparency
  • user perception

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Rationalization: A Neural Machine Translation Approach to Generating Natural Language Explanations'. Together they form a unique fingerprint.

Cite this