DC Field | Value | Language |
---|---|---|
dc.contributor.author | Lee, Jongmin | ko |
dc.contributor.author | Jang, Youngsoo | ko |
dc.contributor.author | Poupart, Pascal | ko |
dc.contributor.author | Kim, Kee-Eung | ko |
dc.date.accessioned | 2017-08-16T08:49:53Z | - |
dc.date.available | 2017-08-16T08:49:53Z | - |
dc.date.created | 2017-06-21 | - |
dc.date.created | 2017-06-21 | - |
dc.date.created | 2017-06-21 | - |
dc.date.issued | 2017-08-24 | - |
dc.identifier.citation | 26th International Joint Conference on Artificial Intelligence, pp.2088 - 2095 | - |
dc.identifier.uri | http://hdl.handle.net/10203/225309 | - |
dc.description.abstract | In this paper, we consider the safe learning scenario where we need to restrict the exploratory behavior of a reinforcement learning agent. Specifically, we treat the problem as a form of Bayesian reinforcement learning in an environment that is modeled as a constrained MDP (CMDP) where the cost function penalizes undesirable situations. We propose a model-based Bayesian reinforcement learning (BRL) algorithm for such an environment, eliciting risk-sensitive exploration in a principled way. Our algorithm efficiently solves the constrained BRL problem by approximate linear programming, and generates a finite state controller in an offline manner. We provide theoretical guarantees and demonstrate empirically that our approach outperforms the state of the art. | - |
dc.language | English | - |
dc.publisher | International Joint Conferences on Artificial Intelligence Organization (IJCAI) | - |
dc.title | Constrained Bayesian Reinforcement Learning via Approximate Linear Programming | - |
dc.type | Conference | - |
dc.identifier.wosid | 000764137502029 | - |
dc.identifier.scopusid | 2-s2.0-85031918650 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 2088 | - |
dc.citation.endingpage | 2095 | - |
dc.citation.publicationname | 26th International Joint Conference on Artificial Intelligence | - |
dc.identifier.conferencecountry | AT | - |
dc.identifier.conferencelocation | Melbourne Convention and Exhibition Center | - |
dc.contributor.localauthor | Kim, Kee-Eung | - |
dc.contributor.nonIdAuthor | Jang, Youngsoo | - |
dc.contributor.nonIdAuthor | Poupart, Pascal | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.