Explaining Recommendations Generated by MDPs

Speaker: Omar Zia Khan

There has been little work in explaining recommendations generated by Markov Decision Processes (MDPs). We analyze the difficulty of explaining policies computed automatically and identify a set of templates that can be used to generate explanations automatically at run-time. These templates are domain-independent and can be used in any application of an MDP. We show that no additional effort is required from the MDP designer for producing such explanations. We use the problem of advising undergraduate students in their course selection to explain the recommendation for selecting specific courses to students. We also propose an extension to leverage domain-specific constructs using ontologies so that explanations can be made more user-friendly.