Actual Causality
Overview
Actual causality asks which events made a particular outcome happen, rather than merely predicting outcomes or finding correlations. Common approaches use Halpern–Pearl definitions in structural causal models, testing counterfactual interventions and refining but-for tests to handle preemption and overdetermination. These methods support explanations and responsibility judgments in settings such as image classifiers, system failures, and multi-agent decisions, with algorithms increasingly designed to find minimal causes at practical scales. A central tension is balancing formal properties and computational efficiency while ensuring that model-based attributions capture relevant dependencies and match intuitive or human judgments.
Latest papers
18 papers4 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.
From Concepts to Actual Causes: Toward Closing the Concept-Malfunction Gap in Robot Foundation Models
A Hybrid Framework for Explainable Aviation Risk: Causal Chain Extraction from Safety Reports with Physics-Based Failure Quantification and Interventional Analysis
Causal Analysis of Failures of Cyber-Physical Systems with Continuous Inputs
Slaying the Hydra: Interaction-Aware Circuit Discovery in Language Models
A Computationally Feasible Framework for Causal Probabilistic Explanation
Responsibility in Multi-Agent Sequential Decision-Making: Comparing Human Judgments to Formal Models of Causal Attribution
Computing Actual Causes for Neural Network Predictions under Structured Causal Inputs
Actual causality in fault trees
Beyond But-for Test: Counterfactual Explanation in Abstract Argumentation via Actual Causality (Extended Version)
Difference-Making without Making a Difference
Explaining Failures of Cyber-Physical Systems with Actual Causality
Neural Information Causality: A Random-Access Capacity Law for Representation Learning
Reconciling Consistency-Based Diagnosis with Actual-Causality-Based Explanations
From Plausible to Causal: Counterfactual Semantics for Policy Evaluation in Simulated Online Communities
policy wind tunnels'' where governance interventions are tested before deployment. But believability is not causality. Claims like intervention reduces escalation'' require causal semantics that current simulation work typically does not specify. We propose adopting the causal counterfactual framework, distinguishing necessary causation (would the outcome have occurred without the intervention?) from sufficient causation (does the intervention reliably produce the outcome?). This distinction maps onto different stakeholder needs: moderators diagnosing incidents require evidence about necessity, while platform designers choosing policies require evidence about sufficiency. We formalize this mapping, show how simulation design can support estimation under explicit assumptions, and argue that the resulting quantities should be interpreted as simulator-conditional causal estimates whose policy relevance depends on simulator fidelity. Establishing this framework now is essential: it helps define what adequate fidelity means and moves the field from simulations that look realistic toward simulations that can support policy changes.