On Counterfactual Explanations under Predictive Multiplicity

Abstract

Counterfactual explanations are usually obtained by identifying the smallest change made to an input to change a prediction made by a fixed model (hereafter called sparse methods). Recent work, however, has revitalized an old insight: there often does not exist one superior solution to a prediction problem with respect to commonly used measures of interest (e.g. error rate). In fact, often multiple different classifiers give almost equal solutions. This phenomenon is known as predictive multiplicity (Breiman, 2001; Marx et al., 2019). In this work, we derive a general upper bound for the costs of counterfactual explanations under predictive multiplicity. Most notably, it depends on a discrepancy notion between two classifiers, which describes how differently they treat negatively predicted individuals. We then compare sparse and data support approaches empirically on real-world data. The results show that data support methods are more robust to multiplicity of different models. At the same time, we show that those methods have provably higher cost of generating counterfactual explanations under one fixed model. In summary, our theoretical and empiricaln results challenge the commonly held view that counterfactual recommendations should be sparse in general.

References in corpus (9)

• Auto-Encoding Variational Bayes
• Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
• Actionable Recourse in Linear Classification
• The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons
• Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers
• Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems
• Algorithmic Recourse: from Counterfactual Explanations to Interventions
• Inverse Classification for Comparison-based Interpretability in Machine Learning
• Predictive Multiplicity in Classification

Cited by in corpus (9)

• A survey of algorithmic recourse: definitions, formulations, solutions, and prospects
• Towards Robust and Reliable Algorithmic Recourse
• Fairness in Agreement With European Values: An Interdisciplinary Perspective on AI Regulation
• Improvement-Focused Causal Recourse (ICR)
• Setting the Right Expectations: Algorithmic Recourse Over Time
• On the Connection between Game-Theoretic Feature Attributions and Counterfactual Explanations
• Understanding Prediction Discrepancies in Machine Learning Classifiers
• Flexible and Robust Counterfactual Explanations with Minimal Satisfiable Perturbations
• Perfect Counterfactuals in Imperfect Worlds: Modelling Noisy Implementation of Actions in Sequential Algorithmic Recourse