Multiple Model-Based Reinforcement Learning

13 years 4 months ago

Download www.cns.atr.jp

We propose a modular reinforcement learning architecture for non-linear, nonstationary control tasks, which we call multiple model-based reinforcement learning (MMRL). The basic idea is to decompose a complex task into multiple domains in space and time based on the predictability of the environmental dynamics. The 1

Kenji Doya, Kazuyuki Samejima, Ken-ichi Katagiri,

Real-time Traffic

Model-based Reinforcement | Modular Reinforcement | NECO 2002 | Nonstationary Control Tasks |

claim paper

» ModelBased Reinforcement Learning in a Complex Domain

» ModelBased Bayesian Reinforcement Learning in Large Structured Domains

» Automatic Feature Selection for ModelBased Reinforcement Learning in Factored MDPs

» Nonparametric ModelBased Reinforcement Learning

» Scaling ModelBased AverageReward Reinforcement Learning for Product Delivery

» Fuzzy Model Based Recognition of Handwritten Hindi Characters

» Exploiting Multiple Secondary Reinforcers in Policy Gradient Reinforcement Learning

» ModelBased Average Reward Reinforcement Learning

Post Info
More Details (n/a)

Added	22 Dec 2010
Updated	22 Dec 2010
Type	Journal
Year	2002
Where	NECO
Authors	Kenji Doya, Kazuyuki Samejima, Ken-ichi Katagiri, Mitsuo Kawato

Comments (0)

Sciweavers

Multiple Model-Based Reinforcement Learning

Model-based Reinforcement | Modular Reinforcement | NECO 2002 | Nonstationary Control Tasks |

Explore & Download

Productivity Tools

Document Tools

Image Tools

Sciweavers