CAUMEL: A Temporal Logic Based Language for Causal Maps to Explain Agent Behaviors
Causal maps are a powerful tools, used to deal with causal relations between events. They are frequently developed for specific issues such as decision analysis and problems diagnostic. The approach described in this paper underlines their novel utility providing a foundation to explain how agents have done actions. In fact, Multi-Agent Systems (MAS) are considered as complex systems, in which agent actions are affected by several factors as uncertain beliefs, intentions of other agents, high interaction, and the dynamic aspect of the environment. Thus, we believe that it is crucial to elucidate the agent system’s behavior. To address the explanation of agent behaviors, this research presents, summarily, our method to build the causal map that corresponds to observed events during agent activities. Then, it focuses on a formal logic theory to interpret the built causal map, which includes causation between temporally ordered actions.
Keywordscausal map explanation agent behavior temporal logic
Unable to display preview. Download preview PDF.
- 1.Johnson, W.L.: Agents that explain their own actions. In: Proceedings of the Fourth Conference on Computer Generated Forces and Behavioral Representation (1994)Google Scholar
- 2.Sengers, P.: Designing comprehensible agents. In: Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence, IJCAI 1999, pp. 1227–1232 (1999)Google Scholar
- 3.Görz, G., Ludwig, B., Reib, P., Schiemann, B., Seutter, T.: Self-describing agents. In: Multikonferenz Wirtschaftsinformatik. GITO-Verlag, Berlin (2008)Google Scholar
- 5.El Fallah Seghrouchni, A., Haddad, S., Mazouzi, H.: A Formal Study of Interactions in Multi-Agent Systems. In: The Proceeding of the 14th International Conference on Computers and Their Applications (1999)Google Scholar
- 6.Guillermo, V., Jorge, J., Gómez, S., Juan, A., Botía, B., Juan, P.: Using Semantic Causality Graphs to Validate MAS Models. In: Innovations in Hybrid Intelligent Systems. Advances in Soft Computing, pp. 9–16. Springer, Heidelberg (2007)Google Scholar
- 7.Vasseur, A.: Dynamic AOP and runtime weaving for Java-How does AspectWerkz address it? In: DAW: Dynamic Aspects Workshop (2004)Google Scholar
- 8.Hedhili, A., Lejouad Chaari, W., Ghédira, K.: Explanation Issue in Multi-Agent Systems. In: The International Conference on Computational Science and Information Management (2012)Google Scholar
- 10.Hedhili Sbaï, A., Lejouad Chaari, W.: Extended Causal Map for Explanation in Multi-Agent Systems. Special Issue on: Intelligent Systems and Applications Using Knowledge and Agent-Based Technologies of the Int. Journal of Intelligent Systems Technologies and Applications 12(3/4), 301–315 (2013)Google Scholar
- 11.Rao, A.S., Georgeff, M.P.: Modeling rational agents within a BDI-architecture. In: Allen, J., Fikes, R., Sandewal, E. (eds.) Proceedings of the Second International Conference on Principles of Knowledge Representation and Reasoning, pp. 473–484. Morgan Kaufman (1991)Google Scholar
- 13.Hedhili Sbaï, A., Lejouad Chaari, W., Ghédira, K.: Intra-agent Explanation Using Temporal and Extended Causal Maps. In: Proceeding of the International Conference Knowledge-Based and Intelligent Information and Engineering Systems, KES 2013, Procedia Computer Science, vol. 22, pp. 241–249 (2013)Google Scholar
- 15.Allen, J.: Towards a general theory of actions and time. Artificial Intelligence, 123–154 (1984)Google Scholar
- 17.Baral, C., Geffond, M.: Reasoning about effects of concurrent actions. Journal of Logic Programming (1997)Google Scholar