{"type":"ScholarlyArticle","title":"α-min: A Compact Approximate Solver For Finite-Horizon POMDPs","authors":[{"name":"Yann Dujardin","slug":"yann-dujardin","given":"Yann","family":"Dujardin","affiliation":""},{"name":"Thomas G. Dietterich","slug":"thomas-g-dietterich","given":"Thomas G.","family":"Dietterich","affiliation":""},{"name":"Iadine Chades","slug":"iadine-chades","given":"Iadine","family":"Chades","affiliation":""}],"abstract":"In many POMDP applications in computational sustainability, it is important that the computed policy have a simple description, so that it can be easily interpreted by stakeholders and decision makers. One measure of simplicity for POMDP value functions is the number of alpha-vectors required to represent the value function. Existing POMDP methods seek to optimize the accuracy of the value function, which can require a very large number of alpha-vectors. This paper studies methods that allow the user to explore the tradeoff between the accuracy of the value function and the number of alpha-vectors. Building on previous point-based POMDP solvers, this paper introduces a new algorithm (alpha-min) that formulates a Mixed Integer Linear Program (MILP) to calculate approximate solutions for finite-horizon POMDP problems with limited numbers of alpha-vectors. At each time-step, alpha-min calculates alpha-vectors to greedily minimize the gap between current upper and lower bounds of the value function. In doing so, good upper and lower bounds are quickly reached allowing a good approximation of the problem with few alpha-vectors. Experimental results show that alpha-min provides good approximate solutions given a fixed number of alpha-vectors on small benchmark problems, on a larger randomly generated problem, as well as on a computational sustainability problem to best manage the endangered Sumatran tiger.","doi":null,"url":"https://ijcai.org/proceedings/2015/366","pdf":"https://ijcai.org/proceedings/2015/0366.pdf","markdown":"https://ijcai.org/proceedings/2015/366.md","bibtex":"https://ijcai.org/proceedings/2015/bibtex/366","volume":{"slug":"2015","title":"Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2015","url":"https://ijcai.org/proceedings/2015/","year":2015,"location":"Buenos Aires, Argentina","dates":"25–31 July 2015","editors":"Qiang Yang and Michael Wooldridge","isbn":"978-1-57735-738-4","publisher":"International Joint Conferences on Artificial Intelligence","doi_prefix":""},"section":"Special Track on Computational Sustainability","track":"Special Track on Computational Sustainability","number":366,"pages":{"first":2582,"last":2587},"keywords":[],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}