Abstract
Responsibility allocation, determining the extent to which agents causally contribute to outcomes, is a fundamental challenge in the design and analysis of multi-agent systems. In this work, we model such systems as concurrent stochastic multi-
player games and introduce a notion of retrospective (backward) counterfactual responsibility, which quantifies an agent’s causal contribution to outcomes resulting from a given strategy profile. To allocate responsibility among agents, we utilise the
Shapley value and formally show that this method satisfies key desirable properties, including fairness and consistency. Building on this foundation, we propose a formal framework that supports both verification and strategic reasoning in responsibility-
aware multi-agent systems. Furthermore, by adopting Nash equilibrium as the solution concept, we demonstrate how to compute stable strategy profiles in which agents trade off responsibility against expected reward.
player games and introduce a notion of retrospective (backward) counterfactual responsibility, which quantifies an agent’s causal contribution to outcomes resulting from a given strategy profile. To allocate responsibility among agents, we utilise the
Shapley value and formally show that this method satisfies key desirable properties, including fairness and consistency. Building on this foundation, we propose a formal framework that supports both verification and strategic reasoning in responsibility-
aware multi-agent systems. Furthermore, by adopting Nash equilibrium as the solution concept, we demonstrate how to compute stable strategy profiles in which agents trade off responsibility against expected reward.
| Original language | English |
|---|---|
| Title of host publication | IJCAI-ECAI 2026, the 35th International Joint Conference on Artificial Intelligence |
| Publisher | International Joint Conferences on Artificial Intelligence |
| Publication status | Accepted/In press - 30 Apr 2026 |
| Event | 35th International Joint Conference on Artificial Intelligence - Bremen, Germany Duration: 15 Aug 2026 → 21 Aug 2026 https://2026.ijcai.org/ |
Publication series
| Name | |
|---|---|
| ISSN (Print) | 1045-0823 |
Conference
| Conference | 35th International Joint Conference on Artificial Intelligence |
|---|---|
| Abbreviated title | IJCAI-ECAI 2026 |
| Country/Territory | Germany |
| City | Bremen |
| Period | 15/08/26 → 21/08/26 |
| Internet address |
Bibliographical note
Chunyan Mu acknowledges the support provided by the University of Aberdeen through the research leave project on “Responsibility-aware Trustworthy AI”, and the National Decommissioning Centre (NDC) project on “Decision-Making for Oil & Gas Decommissioning: A Formal AI-Driven Approach”.Fingerprint
Dive into the research topics of 'Counterfactual Reasoning for Causal Responsibility Attribution in Probabilistic Multi-Agent Systems'. Together they form a unique fingerprint.Cite this
- APA
- Standard
- Harvard
- Vancouver
- Author
- BIBTEX
- RIS