Actionable Recourse in Linear Classification

Abstract

Machine learning models are increasingly used to automate decisions that affect humans - deciding who should receive a loan, a job interview, or a social service. In such applications, a person should have the ability to change the decision of a model. When a person is denied a loan by a credit score, for example, they should be able to alter its input variables in a way that guarantees approval. Otherwise, they will be denied the loan as long as the model is deployed. More importantly, they will lack the ability to influence a decision that affects their livelihood. In this paper, we frame these issues in terms of recourse, which we define as the ability of a person to change the decision of a model by altering actionable input variables (e.g., income vs. age or marital status). We present integer programming tools to ensure recourse in linear classification problems without interfering in model development. We demonstrate how our tools can inform stakeholders through experiments on credit scoring problems. Our results show that recourse can be significantly affected by standard practices in model development, and motivate the need to evaluate recourse in practice.

Extended version. ACM Conference on Fairness, Accountability and Transparency [FAT2019]

References in corpus (9)

• Stealing Machine Learning Models via Prediction APIs
• 'It's Reducing a Human Being to a Percentage'; Perceptions of Justice in Algorithmic Decisions
• Actionable Recourse in Linear Classification
• Supersparse Linear Integer Models for Optimized Medical Scoring Systems
• The Disparate Effects of Strategic Manipulation
• Residual Unfairness in Fair Machine Learning from Prejudiced Data
• Human Perceptions of Fairness in Algorithmic Decision Making: A Case Study of Criminal Risk Prediction
• Maximally Invariant Data Perturbation as Explanation
• Supervising Feature Influence

Cited by in corpus (135)

• Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications
• Deep Neural Networks and Tabular Data: A Survey
• Interpretable Machine Learning -- A Brief History, State-of-the-Art and Challenges
• Actionable Recourse in Linear Classification
• FACE: Feasible and Actionable Counterfactual Explanations
• AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Bias
• The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons
• CERTIFAI: Counterfactual Explanations for Robustness, Transparency, Interpretability, and Fairness of Artificial Intelligence models
• Multi-Objective Counterfactual Explanations
• Learning Model-Agnostic Counterfactual Explanations for Tabular Data
• Problems with Shapley-value-based explanations as feature importance measures
• Beyond Expertise and Roles: A Framework to Characterize the Stakeholders of Interpretable Machine Learning and their Needs
• Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems
• To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods
• AI, Opacity, and Personal Autonomy
• Model-Agnostic Counterfactual Explanations for Consequential Decisions
• Towards Unifying Feature Attribution and Counterfactual Explanations: Different Means to the Same End
• Algorithmic recourse under imperfect causal knowledge: a probabilistic approach
• Algorithmic Recourse: from Counterfactual Explanations to Interventions
• Explainability in Music Recommender Systems
• Explainable Machine Learning for Public Policy: Use Cases, Gaps, and Research Directions
• Counterfactual Shapley Additive Explanations
• Algorithmic Risk Assessments Can Alter Human Decision-Making Processes in High-Stakes Government Contexts
• FlipTest: Fairness Testing via Optimal Transport
• The Conflict Between Explainable and Accountable Decision-Making Algorithms
• A survey of algorithmic recourse: definitions, formulations, solutions, and prospects
• Directive Explanations for Monitoring the Risk of Diabetes Onset: Introducing Directive Data-Centric Explanations and Combinations to Support What-If Explorations
• Causality Learning: A New Perspective for Interpretable Machine Learning
• Getting a CLUE: A Method for Explaining Uncertainty Estimates
• Machine Learning Explainability for External Stakeholders
• Towards Robust and Reliable Algorithmic Recourse
• Recommendations and User Agency: The Reachability of Collaboratively-Filtered Information
• Evaluating Explanation Without Ground Truth in Interpretable Machine Learning
• What-is and How-to for Fairness in Machine Learning: A Survey, Reflection, and Perspective
• When and How to Fool Explainable Models (and Humans) with Adversarial Examples
• ReLAX: Reinforcement Learning Agent eXplainer for Arbitrary Predictive Models
• Generating collective counterfactual explanations in score-based classification via mathematical optimization
• Optimal Decision Making Under Strategic Behavior
• Beyond Individualized Recourse: Interpretable and Interactive Summaries of Actionable Recourses
• Reliable Post hoc Explanations: Modeling Uncertainty in Explainability
• Strategic Classification is Causal Modeling in Disguise
• On Counterfactual Explanations under Predictive Multiplicity
• Towards a Unified Framework for Fair and Stable Graph Representation Learning
• Model extraction from counterfactual explanations
• Rational Shapley Values
• How Well do Feature Visualizations Support Causal Understanding of CNN Activations?
• Learning Strategy-Aware Linear Classifiers
• GAM Coach: Towards Interactive and User-centered Algorithmic Recourse
• Consequence-aware Sequential Counterfactual Generation
• Equalizing Recourse across Groups
• From Explanation to Recommendation: Ethical Standards for Algorithmic Recourse
• Achieving Diversity in Counterfactual Explanations: a Review and Discussion
• Axes for Sociotechnical Inquiry in AI Research
• ViCE: Visual Counterfactual Explanations for Machine Learning Models
• Decisions, Counterfactual Explanations and Strategic Behavior
• GeCo: Quality Counterfactual Explanations in Real Time
• Health improvement framework for planning actionable treatment process using surrogate Bayesian model
• Towards Directive Explanations: Crafting Explainable AI Systems for Actionable Human-AI Interactions
• Interpretable Machine Learning: Moving From Mythos to Diagnostics
• Uncertainty Estimation and Out-of-Distribution Detection for Counterfactual Explanations: Pitfalls and Solutions
• Improvement-Focused Causal Recourse (ICR)
• DECE: Decision Explorer with Counterfactual Explanations for Machine Learning Models
• Setting the Right Expectations: Algorithmic Recourse Over Time
• A Causal Perspective on Meaningful and Robust Algorithmic Recourse
• Synthesizing explainable counterfactual policies for algorithmic recourse with program synthesis
• CARLA: A Python Library to Benchmark Algorithmic Recourse and Counterfactual Explanation Algorithms
• Not All Datasets Are Born Equal: On Heterogeneous Data and Adversarial Examples
• Counterfactual Explanations for Arbitrary Regression Models
• Generating Interpretable Counterfactual Explanations By Implicit Minimisation of Epistemic and Aleatoric Uncertainties
• Disentangling Influence: Using Disentangled Representations to Audit Model Predictions
• Optimal Counterfactual Explanations in Tree Ensembles
• Exploiting Preference Elicitation in Interactive and User-centered Algorithmic Recourse: An Initial Exploration
• Algorithmic Recourse in the Wild: Understanding the Impact of Data and Model Shifts
• Investigating Decision Boundaries of Trained Neural Networks
• Neither Private Nor Fair: Impact of Data Imbalance on Utility and Fairness in Differential Privacy
• On the Connection between Game-Theoretic Feature Attributions and Counterfactual Explanations
• RoCourseNet: Distributionally Robust Training of a Prediction Aware Recourse Model
• Consistent Counterfactuals for Deep Models
• Synthesizing Action Sequences for Modifying Model Decisions
• Efficient Search for Diverse Coherent Explanations
• People Attribute Purpose to Autonomous Vehicles When Explaining Their Behavior: Insights from Cognitive Science for Explainable AI
• Endogenous Macrodynamics in Algorithmic Recourse
• Incorporating Interpretable Output Constraints in Bayesian Neural Networks
• Optimal Counterfactual Explanations for Scorecard modelling
• The Intriguing Relation Between Counterfactual Explanations and Adversarial Examples
• Show Me How: Benefits and Challenges of Agent-Augmented Counterfactual Explanations for Non-Expert Users
• The Importance of Time in Causal Algorithmic Recourse
• Counterfactual Explanations in Sequential Decision Making Under Uncertainty
• Ordered Counterfactual Explanation by Mixed-Integer Linear Optimization
• Towards the Unification and Robustness of Perturbation and Gradient Based Explanations
• For Better or Worse: The Impact of Counterfactual Explanations' Directionality on User Behavior in xAI
• Exploring Counterfactual Explanations Through the Lens of Adversarial Examples: A Theoretical and Empirical Analysis
• Framing Algorithmic Recourse for Anomaly Detection
• PAC-Learning for Strategic Classification
• Even-Ifs From If-Onlys: Are the Best Semi-Factual Explanations Found Using Counterfactuals As Guides?
• Causal Strategic Linear Regression
• Auditing and Debugging Deep Learning Models via Decision Boundaries: Individual-level and Group-level Analysis
• Flexible and Robust Counterfactual Explanations with Minimal Satisfiable Perturbations
• Counterfactual Explanations Can Be Manipulated
• Multi-Differential Fairness Auditor for Black Box Classifiers
• Time Can Invalidate Algorithmic Recourse
• Leveraging Sparse Linear Layers for Debuggable Deep Networks
• Extracting Incentives from Black-Box Decisions
• Seven challenges for harmonizing explainability requirements
• Explaining Black-Box Models through Counterfactuals
• Navigating Explanatory Multiverse Through Counterfactual Path Geometry
• Fair Normalizing Flows
• Generating Causally Compliant Counterfactual Explanations using ASP
• Recourse for reclamation: Chatting with generative language models
• Credit Scoring for Micro-Loans
• Linear Classifiers that Encourage Constructive Adaptation
• Towards User Guided Actionable Recourse
• Towards Feasible Counterfactual Explanations: A Taxonomy Guided Template-based NLG Method
• RAGUEL: Recourse-Aware Group Unfairness Elimination
• On Two XAI Cultures: A Case Study of Non-technical Explanations in Deployed AI System
• ROBY: Evaluating the Robustness of a Deep Model by its Decision Boundaries
• MCCE: Monte Carlo sampling of realistic counterfactual explanations
• Measure Utility, Gain Trust: Practical Advice for XAI Researcher
• Strategic Classification Made Practical
• Estimation of Counterfactual Interventions under Uncertainties
• Perfect Counterfactuals in Imperfect Worlds: Modelling Noisy Implementation of Actions in Sequential Algorithmic Recourse
• SafeAR: Safe Algorithmic Recourse by Risk-Aware Policies
• CounterNet: End-to-End Training of Prediction Aware Counterfactual Explanations
• Adaptive Adversarial Training Does Not Increase Recourse Costs
• Randomized Classifiers vs Human Decision-Makers: Trustworthy AI May Have to Act Randomly and Society Seems to Accept This
• Algorithmic Recourse in Partially and Fully Confounded Settings Through Bounding Counterfactual Effects
• Stateful Strategic Regression
• A Series of Unfortunate Counterfactual Events: the Role of Time in Counterfactual Explanations
• Reasoning about Counterfactuals and Explanations: Problems, Results and Directions
• Self-learn to Explain Siamese Networks Robustly
• Learning Models for Actionable Recourse
• "A cold, technical decision-maker": Can AI provide explainability, negotiability, and humanity?
• Realistic Counterfactual Explanations via Denial Constraints
• Low-Cost Algorithmic Recourse for Users With Uncertain Cost Functions
• Counterfactual Explanations via Latent Space Projection and Interpolation