Model-based reinforcement learning for partially observable games with sampling-based state estimation.

Model-based reinforcement learning for partially observable games with sampling-based state estimation. Research Abstract Details 

Research Abstract Table of Contents

Jump to the:

Model-based reinforcement learning for partially observable games with sampling-based state estimation. Abstract Text:

Hajime Fujita,Shin Ishii,Hajime Fujita,Shin Ishii,

Games constitute a challenging domain of reinforcement learning (RL) for acquiring strategies because many of them include multiple players and many unobservable variables in a large state space. The difficulty of solving such realistic multiagent problems with partial observability arises mainly from the fact that the computational cost for the estimation and prediction in the whole state space, including unobservable variables, is too heavy. To overcome this intractability and enable an agent to learn in an unknown environment, an effective approximation method is required with explicit learning of the environmental model. We present a model-based RL scheme for large-scale multiagent problems with partial observability and apply it to a card game, hearts. This game is a well-defined example of an imperfect information game and can be approximately formulated as a partially observable Markov decision process (POMDP) for a single learning agent. To reduce the computational cost, we use a sampling technique in which the heavy integration required for the estimation and prediction can be approximated by a plausible number of samples. Computer simulation results show that our method is effective in solving such a difficult, partially observable multiagent problem.

Model-based reinforcement learning for partially observable games with sampling-based state estimation. Publishing Authors By Initials

H Fujita,S Ishii,H Fujita,S Ishii,

For similar abstracts research abstracts see: abstracts research

PUBMED ID PMID:

MEDLINE DATE:

Model-based reinforcement learning for partially observable games with sampling-based state estimation. Journal Published:

PUBLICATION TYPE: Journal Article

Journal: Neural computation

VOLUME: 19

Page Numbers: 3051-87

Journal Abbreviation:

ISSN: 0899-7667

DAY: 21

MONTH: Nov

YEAR: 2007

Model-based reinforcement learning for partially observable games with sampling-based state estimation. Information

Number of References:

LANGUAGE: eng

NlmUniqueID: 9426182

Model-based reinforcement learning for partially observable games with sampling-based state estimation. Keywords Mesh Terms:

KEYWORDS:

MESH TERMS:

Chemical & Substance for Abstract: Model-based reinforcement learning for partially observable games with sampling-based state estimation. Information

Substance Name:

Registry Number:

Grant and Affiliation Information for Model-based reinforcement learning for partially observable games with sampling-based state estimation.

AFFILIATION: hajime-f@is.naist.jp

Country: United States

AGENCY:

GRANT:

ACRONYM:

MEDLINETA: Neural Comput

REFSOURCE:

DATABASENAME:

ACCESSION NUMBER:

Number Hits: 0

Model-based reinforcement learning for partially observable games with sampling-based state estimation Related Publications

• A robust, target-driven, cell-based assay for checkpoint kinase 1 inhibitors.
• A selective cellular screening assay for B-Raf and c-Raf kinases.
• Active-site-specific chaperone therapy for Fabry disease. Yin and Yang of enzyme inhibitors.
• Adult cecoanal intussusception caused by cecum cancer: report of a case.
• Characterization of polymorphs and solid-state reactions for paramagnetic systems by 13C solid-state NMR and ab initio calculations.
• Differential requirements for OX40 signals on generation of effector and central memory CD4+ T cells.
• Elucidating connectivity and metal-binding structures of unlabeled paramagnetic complexes by 13C and 1H solid-state NMR under fast magic angle spinning.
• Evaluation of distribution of adenosine A2A receptors in normal human brain measured with [11C]TMSX PET.
• Fringe counting method for synthetic phase with frequency-modulated laser diodes.
• Glucose hypermetabolism in the thalamus of patients with essential blepharospasm.
• Glycine inhibits melanogenesis in vitro and causes hypopigmentation in vivo.
• Historical Survey and New Opportunities in the VUV.
• Magnet lattice for the Siam Photon Source.
• Model-based reinforcement learning for partially observable games with sampling-based state estimation.
• Molecular weights and subunit structure of LamB proteins.
• Outcome of risk-based therapy for infant acute lymphoblastic leukemia with or without an MLL gene rearrangement, with emphasis on late effects: a final report of two consecutive studies, MLL96 and MLL98, of the Japan Infant Leukemia Study Group.
• Overexpression of the granulocyte colony-stimulating factor gene impairs bone morphogenetic protein responsiveness in mice.
• RGS18 acts as a negative regulator of osteoclastogenesis by modulating the acid-sensing OGR1/NFAT signaling pathway.
• Real-time Continuous Estimation of Respiratory Frequency during Sleep based on Heart Rate Time Series.
• Reinforcement learning for a biped robot based on a CPG-actor-critic method.
• Role of oxidative stress from mitochondria on aging and cancer.
• Sensitivity enhancement in (13)C solid-state NMR of protein microcrystals by use of paramagnetic metal ions for optimizing (1)H T(1) relaxation.
• Stimulation of prostanoid IP and EP(2) receptors dilates retinal arterioles and increases retinal and choroidal blood flow in rats.
• Synthesis and reactivities of a bis(cyanamido)-capped triruthenium complex.
• Synthesis of substituted furoates from acrylates and aldehydes by Pd(OAc)(2)/HPMoV/CeCl(3)/O(2) system.
• The perforator pedicled propeller (PPP) flap method: a report of two cases.
• The siam photon laboratory.
• [Development of an algorithm for the detection of early signs of cerebral ischemia on CT images]
• [Evaluation of quality of life for patient with chronic heart failure according to the guidelines of the American College of Cardiology and American Heart Association]
• [The examination of optimal positron emitting nuclide in the estimation of attainment depth within the body of RI beams by positron camera]