{"type":"ScholarlyArticle","title":"Hidden Parameter Markov Decision Processes: A Semiparametric Regression Approach for Discovering Latent Task Parametrizations","authors":[{"name":"Finale Doshi-Velez","slug":"finale-doshi-velez","given":"Finale","family":"Doshi-Velez","affiliation":""},{"name":"George Dimitri Konidaris","slug":"george-dimitri-konidaris","given":"George Dimitri","family":"Konidaris","affiliation":""}],"abstract":"Control applications often feature tasks with similar, but not identical, dynamics. We introduce the Hidden Parameter Markov Decision Process (HiP-MDP), a framework that parametrizes a family of related dynamical systems with a low-dimensional set of latent factors, and introduce a semiparametric regression approach for learning its structure from data. We show that a learned HiP-MDP rapidly identifies the dynamics of new task instances in several settings, flexibly adapting to task variation.","doi":null,"url":"https://ijcai.org/proceedings/2016/206","pdf":"https://ijcai.org/proceedings/2016/0206.pdf","markdown":"https://ijcai.org/proceedings/2016/206.md","bibtex":"https://ijcai.org/proceedings/2016/bibtex/206","volume":{"slug":"2016","title":"Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2016","url":"https://ijcai.org/proceedings/2016/","year":2016,"location":"New York, NY, USA","dates":"9–15 July 2016","editors":"Subbarao Kambhampati","isbn":"978-1-57735-770-4; 978-1-57735-771-1","publisher":"AAAI Press / International Joint Conferences on Artificial Intelligence","doi_prefix":""},"section":null,"track":"","number":206,"pages":{"first":1432,"last":1439},"keywords":[],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}