{"pk":30180,"title":"Value-of-Information based Arbitration between Model-based and Model-freeControl","subtitle":null,"abstract":"There have been numerous attempts in explaining the general learning behaviours using model-based and model-freemethods. While the model-based control is flexible yet computationally expensive in planning, the model-free control isquick but inflexible. Multiple arbitration schemes have been suggested to achieve the data efficiency and computationalefficiency of model-based and model-free control schemes, respectively. In this context, we propose a quantitative ’value-of-information’ based arbitration between both the controllers in order to establish a general computational frameworkfor skill learning. The interacting model-based and model-free reinforcement learning processes are arbitrated using anuncertainty-based value-of-information estimation. We further show that our algorithm performs better than Q-learning aswell as Q-learning with experience replay.","language":"eng","license":{"name":"","short_name":"","text":null,"url":""},"keywords":[],"section":"Member Abstracts, appearing in proceedings only","is_remote":true,"remote_url":"https://escholarship.org/uc/item/9z50z453","frozenauthors":[{"first_name":"Krishn","middle_name":"","last_name":"Bera","name_suffix":"","institution":"International Institute of Information Technology, Hyderabad","department":""},{"first_name":"Yash","middle_name":"","last_name":"Mandilwar","name_suffix":"","institution":"International Institute of Information Technology, Hyderabad","department":""},{"first_name":"Anuj","middle_name":"","last_name":"Shukla","name_suffix":"","institution":"International Institute of Information Technology, Hyderabad","department":""},{"first_name":"Raju","middle_name":"","last_name":"BapiIIIT Hyderabad, Hyderabad,","name_suffix":"","institution":"International Institute of Information Technology, Hyderabad","department":""}],"date_submitted":null,"date_accepted":null,"date_published":"2020-01-01T18:00:00Z","render_galley":null,"galleys":[{"label":"PDF","type":"pdf","path":"https://journalpub.escholarship.org/cognitivesciencesociety/article/30180/galley/20034/download/"}]}