Are contrastive explanations useful?

James Forrest, Somayajulu Sripada, Wei Pang, George M. Coghill

Research output: Contribution to conferenceUnpublished paperpeer-review

Abstract

From the user perspective (data subjects and data controllers), useful explanations of ML decisions are selective, contrastive and social. In this paper, we describe an algorithm for generating selective and contrastive explanations and experimentally study its usefulness to users.

Original languageEnglish
Pages9-16
Number of pages8
Publication statusPublished - 1 Jun 2021
Event2021 SICSA eXplainable Artifical Intelligence Workshop, SICSA XAI 2021 - Aberdeen, United Kingdom
Duration: 1 Jun 2021 → …

Conference

Conference2021 SICSA eXplainable Artifical Intelligence Workshop, SICSA XAI 2021
Country/TerritoryUnited Kingdom
CityAberdeen
Period1/06/21 → …

Keywords

  • Contrastive explanations
  • Interpretable ML
  • XAI

Fingerprint

Dive into the research topics of 'Are contrastive explanations useful?'. Together they form a unique fingerprint.

Cite this