Natural Counterfactuals With Necessary Backtracking

2024/02/02 by Hao, Guang-Yuan, Zhang, Jiji, Huang, Biwei +2
#Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Methodology (stat.ME) #Neural and Evolutionary Computing (cs.NE)

paper · doi:10.48550/arxiv.2402.01607

Abstract

Counterfactual reasoning is pivotal in human cognition and especially important for providing explanations and making decisions. While Judea Pearl's influential approach is theoretically elegant, its generation of a counterfactual scenario often requires too much deviation from the observed scenarios to be feasible, as we show using simple examples. To mitigate this difficulty, we propose a framework of natural counterfactuals and a method for generating counterfactuals that are more feasible with respect to the actual data distribution. Our methodology incorporates a certain amount of backtracking when needed, allowing changes in causally preceding variables to minimize deviations from realistic scenarios. Specifically, we introduce a novel optimization framework that permits but also controls the extent of backtracking with a naturalness criterion. Empirical experiments demonstrate the effectiveness of our method. The code is available at https://github.com/GuangyuanHao/natural_counterfactuals.

Related

• Backtracking Counterfactuals
• Deep Backtracking Counterfactuals for Causally Compliant Explanations
• A Framework for Feasible Counterfactual Exploration incorporating Causality, Sparsity and Density
• A New Approach to Backtracking Counterfactual Explanations: A Unified Causal Framework for Efficient Model Interpretability
• Counterfactual Instances Explain Little