Exploring Counterfactual Explanations Through the Lens of Adversarial Examples: A Theoretical and Empirical Analysis

Abstract

As machine learning (ML) models become more widely deployed in high-stakes applications, counterfactual explanations have emerged as key tools for providing actionable model explanations in practice. Despite the growing popularity of counterfactual explanations, a deeper understanding of these explanations is still lacking. In this work, we systematically analyze counterfactual explanations through the lens of adversarial examples. We do so by formalizing the similarities between popular counterfactual explanation and adversarial example generation methods identifying conditions when they are equivalent. We then derive the upper bounds on the distances between the solutions output by counterfactual explanation and adversarial example generation methods, which we validate on several real-world data sets. By establishing these theoretical and empirical similarities between counterfactual explanations and adversarial examples, our work raises fundamental questions about the design and development of existing counterfactual explanation algorithms.

References in corpus (16)

• Axiomatic Attribution for Deep Networks
• Poisoning Attacks against Support Vector Machines
• Actionable Recourse in Linear Classification
• Generating Natural Adversarial Examples
• Compressed Sensing using Generative Models
• The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons
• Learning Model-Agnostic Counterfactual Explanations for Tabular Data
• Houdini: Fooling Deep Structured Prediction Models
• Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems
• Algorithmic recourse under imperfect causal knowledge: a probabilistic approach
• Identity Matters in Deep Learning
• Imperceptible Adversarial Attacks on Tabular Data
• A survey of algorithmic recourse: definitions, formulations, solutions, and prospects
• Semantics and explanation: why counterfactual explanations produce adversarial examples in deep neural networks
• Adversarial Attacks for Tabular Data: Application to Fraud Detection and Imbalanced Data
• Adversarial Poisoning Attacks and Defense for General Multi-Class Models Based On Synthetic Reduced Nearest Neighbors