Rational Shapley Values

Abstract

Explaining the predictions of opaque machine learning algorithms is an important and challenging task, especially as complex models are increasingly used to assist in high-stakes decisions such as those arising in healthcare and finance. Most popular tools for post-hoc explainable artificial intelligence (XAI) are either insensitive to context (e.g., feature attributions) or difficult to summarize (e.g., counterfactuals). In this paper, I introduce $\text{rational Shapley values}$, a novel XAI method that synthesizes and extends these seemingly incompatible approaches in a rigorous, flexible manner. I leverage tools from decision theory and causal modeling to formalize and implement a pragmatic approach that resolves a number of known challenges in XAI. By pairing the distribution of random variables with the appropriate reference class for a given explanation task, I illustrate through theory and experiments how user goals and knowledge can inform and constrain the solution set in an iterative fashion. The method compares favorably to state of the art XAI tools in a range of quantitative and qualitative comparisons.

arXiv commentTo be presented at the 2022 ACM FAccT Conference

References in corpus (24)

• Least Angle Regression
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Discussion of "Least angle regression" by Efron et al
• Rejoinder to "Least angle regression" by Efron et al
• Interpretable machine learning: definitions, methods, and applications
• Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
• All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously
• The What-If Tool: Interactive Probing of Machine Learning Models
• Actionable Recourse in Linear Classification
• Invariant Risk Minimization
• FACE: Feasible and Actionable Counterfactual Explanations
• Feature relevance quantification in explainable AI: A causal problem
• Problems with Shapley-value-based explanations as feature importance measures
• Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers
• Algorithmic Recourse: from Counterfactual Explanations to Interventions
• Statistical Causality from a Decision-Theoretic Perspective
• A survey of algorithmic recourse: definitions, formulations, solutions, and prospects
• Decision-theoretic foundations for statistical causality

Cited by in corpus (2)

• Counterfactual Shapley Additive Explanations
• Error Analysis of Shapley Value-Based Model Explanations: An Informative Perspective