Towards an argumentation-based approach to explainable planning

Providing transparency of AI planning systems is crucial for their success in practical applications. In order to create a transparent system, a user must be able to query it for explanations about its outputs. We argue that a key underlying principle for this is the use of causality within a planning model, and that argumentation frameworks provide an intuitive representation of such causality. In this paper, we discuss how argumentation can aid in extracting causalities from plans and models, and how they can create explanations from them.

[1]  Michael Wooldridge,et al.  Properties and Complexity of Some Formal Inter-agent Dialogues , 2003, J. Log. Comput..

[2]  Yu Zhang,et al.  Plan Explanations as Model Reconciliation: Moving Beyond Explanation as Soliloquy , 2017, IJCAI.

[3]  Lukás Chrpa,et al.  Towards Getting Domain Knowledge: Plans Analysis through Investigation of Actions Dependencies , 2008, FLAIRS.

[4]  Denis J. Hilton,et al.  The course of events: counterfactuals, causal sequences, and explanation , 2005 .

[5]  Alexander Bochman.  Propositional Argumentation and Causal Reasoning , 2005, IJCAI.

[6]  Malte Helmert,et al.  A Planning Heuristic Based on Causal Graph Analysis , 2004, ICAPS.

[7]  Tim Miller,et al.  Contrastive explanation: a structural-model approach , 2018, The Knowledge Engineering Review.

[8]  Elizabeth Sklar,et al.  Argumentation-based dialogue games for shared control in human-robot systems , 2015, HRI 2015.

[9]  David E. Smith.  Planning as an Iterative Process , 2012, AAAI.

[10]  Anders Jonsson,et al.  The Complexity of Planning Problems With Simple Causal Graphs , 2008, J. Artif. Intell. Res..

[11]  Maria Fox,et al.  Explainable Planning , 2017, ArXiv.

[12]  Henry Prakken,et al.  A general account of argumentation with preferences , 2013, Artif. Intell..

[13]  Francesca Toni,et al.  Argumentation for Explainable Scheduling , 2019, AAAI.

[14]  Judea Pearl,et al.  Graphical Models for Probabilistic and Causal Reasoning , 1997, The Computer Science and Engineering Handbook.