Explaining Explanations in AI

Proceedings of FAT* ’19: Conference on Fairness, Accountability, and Transparency (FAT* ’19), January 29–31, 2019, Atlanta, GA, USA. ACM, New York, NY, USA, doi/10.1145/3287560.3287574, ISBN: 978-1-4503-6125-5

10 Pages Posted: 27 Nov 2018

See all articles by Brent Mittelstadt

Brent Mittelstadt

University of Oxford - Oxford Internet Institute

Chris Russell

University of Surrey

Sandra Wachter

University of Oxford - Oxford Internet Institute

Date Written: November 4, 2018

Abstract

Recent work on interpretability in machine learning and AI has focused on the building of simplified models that approximate the true criteria used to make decisions. These models are a useful pedagogical device for teaching trained professionals how to predict what decisions will be made by the complex system, and most importantly how the system might break. However, when considering any such model it’s important to remember Box’s maxim that "All models are wrong but some are useful." We focus on the distinction between these models and explanations in philosophy and sociology. These models can be understood as a "do it yourself kit" for explanations, allowing a practitioner to directly answer "what if questions" or generate contrastive explanations without external assistance. Although a valuable ability, giving these models as explanations appears more difficult than necessary, and other forms of explanation may not have the same trade-offs. We contrast the different schools of thought on what makes an explanation, and suggest that machine learning might benefit from viewing the problem more broadly.

Keywords: interpretability, explanations, accountability, philosophy of science, data ethics, machine learning, artificial intelligence, automated decision-making

Suggested Citation

Mittelstadt, Brent and Russell, Chris and Wachter, Sandra, Explaining Explanations in AI (November 4, 2018). Proceedings of FAT* ’19: Conference on Fairness, Accountability, and Transparency (FAT* ’19), January 29–31, 2019, Atlanta, GA, USA. ACM, New York, NY, USA, doi/10.1145/3287560.3287574, ISBN: 978-1-4503-6125-5. Available at SSRN: https://ssrn.com/abstract=3278331

Brent Mittelstadt (Contact Author)

University of Oxford - Oxford Internet Institute ( email )

1 St. Giles
University of Oxford
Oxford OX1 3PG Oxfordshire, Oxfordshire OX1 3JS
United Kingdom

Chris Russell

University of Surrey ( email )

Guildford
Guildford, Surrey GU2 5XH
United Kingdom

Sandra Wachter

University of Oxford - Oxford Internet Institute ( email )

1 St. Giles
University of Oxford
Oxford OX1 3PG Oxfordshire, Oxfordshire OX1 3JS
United Kingdom

Register to save articles to
your library

Register

Paper statistics

Downloads
276
rank
103,712
Abstract Views
1,460
PlumX Metrics