A Computationally Feasible Framework for Causal Probabilistic Explanation
2026-09-03 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors address the challenge of explaining why specific results happen and which factors are responsible, a problem important in science and policy. Current methods either provide exact answers but only for small models, or work on large data but ignore the causal relationships. Their new method, Probabilistic Causal Impact (PCI), combines these approaches by framing explanation as a probability estimation problem, making it practical for complex models while respecting causal reasoning. They test PCI on synthetic and real data, showing it works well across different scenarios, including large-scale machine learning models.
Actual CausalityCounterfactualsCausal ModelsSHAPProbability of NecessityProbability of SufficiencyMonte Carlo MethodsCausal InferenceExplainabilityMachine Learning
Authors
Rafal Urbaniak, Sam Witty, Daniel Waxman, Andy Zane, Poorva Garg, Emily Bunnapradist, Sankaran Vaidyanathan, Jack Feser, Drew Lehe, Eli Bingham
Abstract
Explaining why a specific outcome occurred, and which inputs deserve the blame or credit, is central to philosophical, scientific, and policy analysis. Existing tools split into two camps. The theory of actual causality (AC) gives principled verdicts, but only for toy-sized models, because computing them requires enumerating counterfactual scenarios. Scalable attribution methods like SHAP (or even causal SHAP) at least partially ignore the causal structure that generated the data, and can give answers that conflict with a careful causal analysis. We close this gap with Probabilistic Causal Impact (PCI). PCI builds on actual causality and on Pearl's notions of probability of necessity and sufficiency, but recasts the question of explainability as an estimation problem on a probabilistic causal model that is easily approximated via Monte Carlo. By specifying a distribution over "candidate explanations," a distribution over counterfactual values, and a scoring function, PCI provides tractable, causally grounded, graded explanations, generalizing AC and Pearl's probability of causation as degenerate cases. We evaluate PCI in synthetic and real-world examples, spanning consistency checks with AC, scaling experiments, complex continuous-valued dynamical systems, and a real-world deployed causal machine learning model trained on millions of datapoints.