Sufficient Reasons and Explanations for Reactive Systems

2026-10-01

AI Summary

This project addresses the challenges of temporal causality and interpretability in reactive systems by proposing a general temporal interpretability framework. Methodologically, it unifies the concepts of sufficient reasons and contrastive explanations, extends neural network interpretation techniques to temporal logic specifications, constructs symbolic representations along the temporal dimension, and integrates formal verification with symbolic execution for solving. The primary theoretical contribution lies in establishing the computational complexity boundaries for various forms of temporal explanations. Practically, the effectiveness of the proposed approach is validated through a prototype system. Overall, this work provides a novel paradigm for the trustworthy analysis of complex reactive systems.

Abstract

We address the problem of temporal causality and explainability for reactive systems, and, in this setting, study sufficient reasons and contrastive explanations. These two notions are well-known explainability measures in the context of neural networks. In this work, we unify these notions for reactive systems and formal specifications given in temporal logic, providing dedicated definitions for sufficient reasons and contrastive explanations. We then lift these definitions to \emph{temporal} sufficient reasons and contrastive explanations, providing more general and symbolic representations of explainability. We analyze the complexity of both verifying and finding explanations of the different types, and we demonstrate our approach using a prototype implementation.

Problem

• reactive systems
• temporal causality
• explainability
• sufficient reasons
• contrastive explanations

Innovation

• temporal causality
• reactive systems
• sufficient reasons
• contrastive explanations
• temporal logic