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 | 2017-11-21T03:03:04Z | - |
dc.date.available | 2017-11-21T03:03:04Z | - |
dc.date.created | 2017-11-14 | - |
dc.date.created | 2017-11-14 | - |
dc.date.issued | 2017-09-18 | - |
dc.identifier.citation | Scaling-Up Reinforcement Learning Workshop at ECML PKDD 2017 | - |
dc.identifier.uri | http://hdl.handle.net/10203/227104 | - |
dc.description.abstract | In this paper, we highlight our recent work~\cite{Lee2017} considering 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 (BRL) in an environment that is modeled as a constrained MDP (CMDP) where the cost function penalizes undesirable situations. We propose a model-based 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 off-line manner. We provide theoretical guarantees and demonstrate empirically that our approach outperforms the state of the art. | - |
dc.language | English | - |
dc.publisher | Scaling-Up Reinforcement Learning Workshop at ECML PKDD 2017 | - |
dc.title | Constrained Bayesian Reinforcement Learning via Approximate Linear Programming | - |
dc.type | Conference | - |
dc.identifier.wosid | 000764137502029 | - |
dc.type.rims | CONF | - |
dc.citation.publicationname | Scaling-Up Reinforcement Learning Workshop at ECML PKDD 2017 | - |
dc.identifier.conferencecountry | MA | - |
dc.identifier.conferencelocation | Hotel Aleksandar Palace, Skopje | - |
dc.contributor.localauthor | Kim, Kee-Eung | - |
dc.contributor.nonIdAuthor | Poupart, Pascal | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.