Contrastive Explanation: A Structural-Model Approach

Abstract

This paper presents a model of contrastive explanation using structural casual models. The topic of causal explanation in artificial intelligence has gathered interest in recent years as researchers and practitioners aim to increase trust and understanding of intelligent decision-making. While different sub-fields of artificial intelligence have looked into this problem with a sub-field-specific view, there are few models that aim to capture explanation more generally. One general model is based on structural causal models. It defines an explanation as a fact that, if found to be true, would constitute an actual cause of a specific event. However, research in philosophy and social sciences shows that explanations are contrastive: that is, when people ask for an explanation of an event -- the fact -- they (sometimes implicitly) are asking for an explanation relative to some contrast case; that is, "Why P rather than Q?". In this paper, we extend the structural causal model approach to define two complementary notions of contrastive explanation, and demonstrate them on two classical problems in artificial intelligence: classification and planning. We believe that this model can help researchers in subfields of artificial intelligence to better understand contrastive explanation.

References in corpus (2)

• Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
• Contrastive Explanations for Reinforcement Learning in terms of Expected Consequences

Cited by in corpus (27)

• Knowing About Knowing: An Illusion of Human Competence Can Hinder Appropriate Reliance on AI Systems
• Opening the Black-Box: A Systematic Review on Explainable AI in Remote Sensing
• Sensible AI: Re-imagining Interpretability and Explainability using Sensemaking Theory
• Formalizing Trust in Artificial Intelligence: Prerequisites, Causes and Goals of Human Trust in AI
• Towards Relatable Explainable AI with the Perceptual Process
• Measurable Counterfactual Local Explanations for Any Classifier
• A survey of algorithmic recourse: definitions, formulations, solutions, and prospects
• The Emerging Landscape of Explainable AI Planning and Decision Making
• Towards Explainable AI Planning as a Service
• Reasoning about Causality in Games
• Forms of Understanding for XAI-Explanations
• On Relating 'Why?' and 'Why Not?' Explanations
• Weight of Evidence as a Basis for Human-Oriented Explanations
• A Bayesian Account of Measures of Interpretability in Human-AI Interaction
• Distal Explanations for Model-free Explainable Reinforcement Learning
• Towards Contrastive Explanations for Comparing the Ethics of Plans
• A moving target in AI-assisted decision-making: Dataset shift, model updating, and the problem of update opacity
• Generative Counterfactuals for Neural Networks via Attribute-Informed Perturbation
• COIN: Counterfactual inpainting for weakly supervised semantic segmentation for medical images
• SCOUT: Self-aware Discriminant Counterfactual Explanations
• Recourse for reclamation: Chatting with generative language models
• LEx: A Framework for Operationalising Layers of Machine Learning Explanations
• Explanation format does not matter; but explanations do -- An Eggsbert study on explaining Bayesian Optimisation tasks
• CE-MRS: Contrastive Explanations for Multi-Robot Systems
• XAlgo: a Design Probe of Explaining Algorithms' Internal States via Question-Answering
• Counterfactual Interventions Reveal the Causal Effect of Relative Clause Representations on Agreement Prediction
• Towards Transparent Robotic Planning via Contrastive Explanations