City Research Online

Measurable counterfactual local explanations for any classifier

White, A. and d'Avila Garcez, A. S. (2020). Measurable counterfactual local explanations for any classifier. Frontiers in Artificial Intelligence and Applications, 325, pp. 2529-2535. doi: 10.3233/FAIA200387

Abstract

We propose a novel method for explaining the predictions of any classifier. In our approach, local explanations are expected to explain both the outcome of a prediction and how that prediction would change if athings had been different'. Furthermore, we argue that satisfactory explanations cannot be dissociated from a notion and measure of fidelity, as advocated in the early days of neural networks' knowledge extraction. We introduce a definition of fidelity to the underlying classifier for local explanation models which is based on distances to a target decision boundary. A system called CLEAR: Counterfactual Local Explanations via Regression, is introduced and evaluated. CLEAR generates b-counterfactual explanations that state minimum changes necessary to flip a prediction's classification. CLEAR then builds local regression models, using the b-counterfactuals to measure and improve the fidelity of its regressions. By contrast, the popular LIME method [17], which also uses regression to generate local explanations, neither measures its own fidelity nor generates counterfactuals. CLEAR's regressions are found to have significantly higher fidelity than LIME's, averaging over 40% higher in this paper's five case studies.

Publication Type: Article
Additional Information: © 2020 The authors and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments: School of Mathematics, Computer Science & Engineering > Computer Science
Date available in CRO: 04 May 2021 13:57
Date deposited: 4 May 2021
Date of first online publication: 24 August 2020
URI: https://openaccess.city.ac.uk/id/eprint/25968
[img]
Preview
Text - Published Version
Available under License Creative Commons Attribution Non-commercial.

Download (469kB) | Preview

Export

Downloads

Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login