Model-Agnostic Counterfactual Explanations for Consequential Decisions

Abstract

Predictive models are being increasingly used to support consequential decision making at the individual level in contexts such as pretrial bail and loan approval. As a result, there is increasing social and legal pressure to provide explanations that help the affected individuals not only to understand why a prediction was output, but also how to act to obtain a desired outcome. To this end, several works have proposed optimization-based methods to generate nearest counterfactual explanations. However, these methods are often restricted to a particular subset of models (e.g., decision trees or linear models) and differentiable distance functions. In contrast, we build on standard theory and tools from formal verification and propose a novel algorithm that solves a sequence of satisfiability problems, where both the distance function (objective) and predictive model (constraints) are represented as logic formulae. As shown by our experiments on real-world data, our algorithm is: i) model-agnostic ({non-}linear, {non-}differentiable, {non-}convex); ii) data-type-agnostic (heterogeneous features); iii) distance-agnostic (${\ell }_0,{\ell }_1,{\ell }_{\infty }$, and combinations thereof); iv) able to generate plausible and diverse counterfactuals for any sample (i.e., 100% coverage); and v) at provably optimal distances.

References in corpus (6)

• Scikit-learn: Machine Learning in Python
• Towards A Rigorous Science of Interpretable Machine Learning
• Interpretable machine learning: definitions, methods, and applications
• The What-If Tool: Interactive Probing of Machine Learning Models
• Actionable Recourse in Linear Classification
• Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead

Cited by in corpus (24)

• The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons
• To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods
• Algorithmic Recourse: from Counterfactual Explanations to Interventions
• Towards Robust and Reliable Algorithmic Recourse
• A Framework and Benchmarking Study for Counterfactual Generating Methods on Tabular Data
• Improving Visualization Interpretation Using Counterfactuals
• Beyond Individualized Recourse: Interpretable and Interactive Summaries of Actionable Recourses
• How Well do Feature Visualizations Support Causal Understanding of CNN Activations?
• Decisions, Counterfactual Explanations and Strategic Behavior
• Why model why? Assessing the strengths and limitations of LIME
• Synthesizing explainable counterfactual policies for algorithmic recourse with program synthesis
• DECE: Decision Explorer with Counterfactual Explanations for Machine Learning Models
• Algorithmic Recourse in the Wild: Understanding the Impact of Data and Model Shifts
• An exact counterfactual-example-based approach to tree-ensemble models interpretability
• The Intriguing Relation Between Counterfactual Explanations and Adversarial Examples
• Towards the Unification and Robustness of Perturbation and Gradient Based Explanations
• Extracting Incentives from Black-Box Decisions
• Counterfactual Explanations for Machine Learning: Challenges Revisited
• Navigating Explanatory Multiverse Through Counterfactual Path Geometry
• Perfect Counterfactuals in Imperfect Worlds: Modelling Noisy Implementation of Actions in Sequential Algorithmic Recourse
• Finding Optimal Diverse Feature Sets with Alternative Feature Selection
• SafeAR: Safe Algorithmic Recourse by Risk-Aware Policies
• TrustyAI Explainability Toolkit
• Post-discovery Analysis of Anomalous Subsets