Home  | Publications | MKH+20

General Pitfalls of Model-Agnostic Interpretation Methods for Machine Learning Models

MCML Authors

Christian Alexander Scholbeck

Dr.

* Former Member

→ Group Bernd Bischl
 Statistical Learning and Data Science

Moritz Grosse-Wentrup

Prof. Dr.

Principal Investigator

* Former Principal Investigator

Link to Profile Bernd Bischl PI Matchmaking

Bernd Bischl

Prof. Dr.

Director

Statistical Learning and Data Science

Abstract

An increasing number of model-agnostic interpretation techniques for machine learning (ML) models such as partial dependence plots (PDP), permutation feature importance (PFI) and Shapley values provide insightful model interpretations, but can lead to wrong conclusions if applied incorrectly. We highlight many general pitfalls of ML model interpretation, such as using interpretation techniques in the wrong context, interpreting models that do not generalize well, ignoring feature dependencies, interactions, uncertainty estimates and issues in high-dimensional settings, or making unjustified causal interpretations, and illustrate them with examples. We focus on pitfalls for global methods that describe the average model behavior, but many pitfalls also apply to local methods that explain individual predictions. Our paper addresses ML practitioners by raising awareness of pitfalls and identifying solutions for correct model interpretation, but also addresses ML researchers by discussing open issues for further research.

inproceedings

XXAI @ICML 2020

Workshop on Extending Explainable AI Beyond Deep Models and Classifiers at the 37th International Conference on Machine Learning. Virtual, Jul 12-18, 2020.

Authors

C. Molnar • G. König • J. Herbinger • T. Freiesleben • S. Dandl • C. A. Scholbeck • G. Casalicchio • M. Grosse-Wentrup • B. Bischl

Links

DOI

Research Area

 A1 | Statistical Foundations & Explainability

BibTeXKey: MKH+20

Back to Top