Multi-Objective Counterfactual Explanations

Abstract

Counterfactual explanations are one of the most popular methods to make predictions of black box machine learning models interpretable by providing explanations in the form of `what-if scenarios'. Most current approaches optimize a collapsed, weighted sum of multiple objectives, which are naturally difficult to balance a-priori. We propose the Multi-Objective Counterfactuals (MOC) method, which translates the counterfactual search into a multi-objective optimization problem. Our approach not only returns a diverse set of counterfactuals with different trade-offs between the proposed objectives, but also maintains diversity in feature space. This enables a more detailed post-hoc analysis to facilitate better understanding and also more options for actionable user responses to change the predicted outcome. Our approach is also model-agnostic and works for numerical and categorical input features. We show the usefulness of MOC in concrete cases and compare our approach with state-of-the-art methods for counterfactual explanations.

References in corpus (5)

• One pixel attack for fooling deep neural networks
• OpenML: networked science in machine learning
• Actionable Recourse in Linear Classification
• FACE: Feasible and Actionable Counterfactual Explanations
• Interpretable Predictions of Tree-based Ensembles via Actionable Feature Tweaking

Cited by in corpus (24)

• Interpretable Machine Learning -- A Brief History, State-of-the-Art and Challenges
• Counterfactual Explainable Recommendation
• Towards Unifying Feature Attribution and Counterfactual Explanations: Different Means to the Same End
• Counterfactual Shapley Additive Explanations
• Active learning of the thermodynamics-dynamics tradeoff in protein condensates
• ReLAX: Reinforcement Learning Agent eXplainer for Arbitrary Predictive Models
• Scientific Inference With Interpretable Machine Learning: Analyzing Models to Learn About Real-World Phenomena
• Keep Your Friends Close and Your Counterfactuals Closer: Improved Learning From Closest Rather Than Plausible Counterfactual Explanations in an Abstract Setting
• Let's Go to the Alien Zoo: Introducing an Experimental Framework to Study Usability of Counterfactual Explanations for Machine Learning
• Benchmarking Instance-Centric Counterfactual Algorithms for XAI: From White Box to Black Box
• Consequence-aware Sequential Counterfactual Generation
• Evolutionary approaches to explainable machine learning
• Synthesizing explainable counterfactual policies for algorithmic recourse with program synthesis
• TABCF: Counterfactual Explanations for Tabular Data Using a Transformer-Based VAE
• GradCFA: A Hybrid Gradient-Based Counterfactual and Feature Attribution Explanation Algorithm for Local Interpretation of Neural Networks
• Interpretable Machine Learning for TabPFN
• Interpretable Machine Learning for Survival Analysis
• The Intriguing Relation Between Counterfactual Explanations and Adversarial Examples
• Digital Voodoo Dolls
• Explainable Reinforcement Learning for Formula One Race Strategy
• Even-Ifs From If-Onlys: Are the Best Semi-Factual Explanations Found Using Counterfactuals As Guides?
• Perfect Counterfactuals in Imperfect Worlds: Modelling Noisy Implementation of Actions in Sequential Algorithmic Recourse
• Finding Optimal Diverse Feature Sets with Alternative Feature Selection
• Realistic Counterfactual Explanations via Denial Constraints