City Research Online

Interpreting Deep Learning Models with Marginal Attribution by Conditioning on Quantiles

Merz, M., Richman, R., Tsanakas, A. ORCID: 0000-0003-4552-5532 & Wüthrich, M. (2022). Interpreting Deep Learning Models with Marginal Attribution by Conditioning on Quantiles. Data Mining and Knowledge Discovery,

Abstract

A vast and growing literature on explaining deep learning models has emerged. This paper contributes to that literature by introducing a global gradient-based model-agnostic method, which we call Marginal Attribution by Conditioning on Quantiles (MACQ). Our approach is based on analyzing the marginal attribution of predictions (outputs) to individual features (inputs). Specifically, we consider variable importance by fixing (global) output levels, and explaining how features marginally contribute to these fixed global output levels. MACQ can be seen as a marginal attribution counterpart to approaches such as accumulated local effects (ALE), which study the sensitivities of outputs by perturbing inputs. Furthermore, MACQ allows us to separate marginal attribution of individual features from interaction effects and to visualize the 3-way relationship between marginal attribution, output level, and feature value.

Publication Type: Article
Additional Information: This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://link.springer.com/journal/10618
Publisher Keywords: explainable AI (XAI), model-agnostic tools, deep learning, attribution, accumu- lated local effects (ALE), partial dependence plot (PDP), locally interpretable model-agnostic explanation (LIME), variable importance, post-hoc analysis, interaction
Subjects: H Social Sciences > HB Economic Theory
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
T Technology > T Technology (General)
Departments: Bayes Business School > Actuarial Science & Insurance
[img] Text - Accepted Version
This document is not freely accessible due to copyright restrictions.

To request a copy, please use the button below.

Request a copy

Export

Downloads

Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login