A Survey of Algorithmic Recourse: Contrastive Explanations and Consequential Recommendations

2022/04/04 by Amir-Hossein Karimi, Gilles Barthe, Bernhard Schölkopf +1 · 24 citations
Computer Science · Social Sciences · #Adversarial Robustness in Machine Learning #Explainable Artificial Intelligence (XAI) #Ethics and Social Impacts of AI

paper · pdf · doi:10.1145/3527848

Abstract

Machine learning is increasingly used to inform decision making in sensitive situations where decisions have consequential effects on individuals’ lives. In these settings, in addition to requiring models to be accurate and robust, socially relevant values such as fairness, privacy, accountability, and explainability play an important role in the adoption and impact of said technologies. In this work, we focus on algorithmic recourse , which is concerned with providing explanations and recommendations to individuals who are unfavorably treated by automated decision-making systems. We first perform an extensive literature review, and align the efforts of many authors by presenting unified definitions, formulations, and solutions to recourse. Then, we provide an overview of the prospective research directions toward which the community may engage, challenging existing assumptions and making explicit connections to other ethical challenges such as security, privacy, and fairness.

Citations

• Conversational processes and causal explanation.
• Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author)
• Causes and Explanations: A Structural-Model Approach. Part I: Causes
• The magical number seven, plus or minus two: Some limits on our capacity for processing information.
• Inverse Classification for Comparison-based Interpretability in Machine Learning
• Interpretable Credit Application Predictions With Counterfactual Explanations
• Explainable Artificial Intelligence (XAI) on TimeSeries Data: A Survey
• Discriminative Attribution from Counterfactuals
• Solving the Class Imbalance Problem Using a Counterfactual Method for Data Augmentation

Cited by

• Counterfactual Explanations for Time Series Should be Human-Centered and Temporally Coherent in Interventions
• Causal Explanations for Disparate Trends: Where and Why?
• Imaginative Thought
• Few-Shot Knowledge Distillation of LLMs With Counterfactual Explanations
• Desirable Effort Fairness and Optimality Trade-offs in Strategic Learning
• Synthesising Counterfactual Explanations via Label-Conditional Gaussian Mixture Variational Autoencoders
• Understanding Contestability on the Margins: Implications for the Design of Algorithmic Decision-making in Public Services
• Reinforcement Learning for Durable Algorithmic Recourse
• The Role of Causality in Algorithmic Recourse
• When AI Does the Work, What Is Learning For? Post-Instrumental Learning and the Risk of Capacity Dissolution
• Explainable AI Isn't Enough! Rethinking Algorithmic Contestability
• Transparent and Fair Profiling in Employment Services: Evidence from Switzerland
• Collective Recourse for Generative Urban Visualizations
• Counterfactual Scenarios for Automated Planning
• Informative Post-Hoc Explanations Only Exist for Simple Functions
• Your Recourse, My Loss? Algorithmic Recourse under Shared Constraints
• What If, But Privately: Private Counterfactual Retrieval
• A model-agnostic and data-independent tabu search algorithm to generate counterfactuals for tabular, image, and text data
• A Generalized-Bayes Perspective on Counterfactual Explanations: Posterior-Based Decision-Making and Evaluation
• ReVise: A Human-AI Interface for Incremental Algorithmic Recourse
• Data Science Principles for Interpretable and Explainable AI
• Argumentative Ensembling for Robust Recourse under Model Multiplicity
• Relative Explanations for Contextual Problems with Endogenous Uncertainty: An Application to Competitive Facility Location
• Performative Validity of Recourse Explanations

Related

• A survey of algorithmic recourse: definitions, formulations, solutions,\n and prospects
• Algorithmic Recourse: from Counterfactual Explanations to Interventions
• Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems
• Explanation Hacking: The perils of algorithmic recourse
• Explainable AI Isn't Enough! Rethinking Algorithmic Contestability