A Markov Decision Process (MDP) policy presents, for each state, an action, which preferably maximizes the expected utility accrual over time. In this article, we present a novel explanation system for MDP policies. The system interactively generates conversational English-language explanations of the actions suggested by an optimal policy, and does so in real time. We rely on natural language explanations in order to build trust between the user and the explanation system, leveraging existing research in psychology in order to generate salient explanations. Our explanation system is designed for portability between domains and uses a combination of domain-specific and domain-independent techniques. The system automatically extracts implicit knowledge from an MDP model and accompanying policy. This MDP-based explanation system can be ported between applications without additional effort by knowledge engineers or model builders. Our system separates domain-specific data from the explanation logic, allowing for a robust system capable of incremental upgrades. Domain-specific explanations are generated through case-based explanation techniques specific to the domain and a knowledge base of concept mappings used to generate English-language explanations.
We present an online, heuristic algorithm for learning Conditional Preference networks (CP-nets) from user queries. This is the first efficient and resolute CP-net learning algorithm: if a preference order can be represented as a CP-net, our algorithm learns a CP-net in time n p , where p is a bound on the number of parents a node may have. The learned CP-net is guaranteed to be consistent with the original CP-net on all queries from the learning process. We tested the algorithm on randomly generated CP-nets; the learned CP-nets agree with the originals on a high percent of non-training preference comparisons.
We detail some lessons learned while designing and testing a decision-theoretic advising support tool for undergraduates at a large state university. Between 2009 and 2011 we conducted two surveys of over 500 students in multiple majors and colleges. These surveys asked students detailed questions about their preferences concerning course selection, advising, and career paths. We present data from this study which may be helpful for faculty and staff who advise undergraduate students. We find that advising support software tools can augment the student-advisor relationship, particularly in terms of course planning, but cannot and should not replace in-person advising.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.