Counterfactual Explanations for MITL Violations

2024-11-29

Foundations of Software Technology and Theoretical Computer Science

AI Summary

This work addresses the challenge of root-cause diagnosis for Metric Interval Temporal Logic (MITL) specification violations in real-time systems. We propose the first counterfactual causal analysis framework tailored to networked timed automata. Methodologically, we adapt the Halpern–Pearl causal model to the semantics of timed automata, formally define counterfactual causality for MITL violations, and integrate MITL model checking, symbolic trace analysis, and causal graph construction to jointly attribute discrete actions and real-time delays. Our contributions are threefold: (1) a formal, verifiable definition of counterfactual causality for timed systems; (2) an efficient algorithm for generating minimal counterfactual interventions that restore MITL satisfaction; and (3) empirical evaluation on standard benchmarks demonstrating substantial reduction in manual debugging effort and delivering highly interpretable, causally grounded diagnostic reports.

Technology Category

• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Action, Change, and Causality
• Machine Learning: Causal Learning

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

MITL is a temporal logic that facilitates the verification of real-time systems by expressing the critical timing constraints placed on these systems. MITL specifications can be checked against system models expressed as networks of timed automata. A violation of an MITL specification is then witnessed by a timed trace of the network, i.e., an execution consisting of both discrete actions and real-valued delays between these actions. Finding and fixing the root cause of such a violation requires significant manual effort since both discrete actions and real-time delays have to be considered. In this paper, we present an automatic explanation method that eases this process by computing the root causes for the violation of an MITL specification on the execution of a network of timed automata. This method is based on newly developed definitions of counterfactual causality tailored to networks of timed automata in the style of Halpern and Pearl's actual causality. We present and evaluate a prototype implementation that demonstrates the efficacy of our method on several benchmarks from the literature.

Problem

• Identifies root causes of MITL specification violations
• Automates analysis of timed automata network executions
• Provides counterfactual explanations for real-time system errors

Innovation

• Automatic root cause analysis for MITL violations
• Counterfactual causality for timed automata networks
• Prototype implementation validated on benchmark cases