Intrinsic, Dialogic, and Impact Measures of Success for Explainable AI

Activity: Talk or presentationOral presentationscience to science / art to art

Description

This paper presents a brief overview of requirements for development and evaluation of human centered explainable systems. We propose three perspectives on evaluation models for explainable AI that include intrinsic measures, dialogic measures and impact measures. The paper outlines these different perspectives and looks at how the separation might be used for explanation evaluation bench marking and integration into design and development. We propose several avenues for future work.
Period2021
Event titleModeling and Reasoning in Context Workshop: Human-Centric and Contextual Systems
Event typeOnline-Workshop
Conference number12th
LocationMontreal, CanadaShow on map

Keywords

  • Artificial intelligence
  • Explainable AI
  • Evaluation

Fields of Science and Technology Classification 2012

  • 102 Computer Sciences