bibtype	C - Conference Paper (international conference)
ARLID	0503817
utime	20241106135724.9
mtime	20190408235959.9
SCOPUS	85064837601
DOI	10.5220/0007587208570864
title (primary) (eng)	Balancing Exploitation and Exploration via Fully Probabilistic Design of Decision Policies
specification	page_count 8 s. media_type P
serial	ARLID cav_un_epca*0503816 ISBN 978-989-758-350-6 title Proceedings of the 11th International Conference on Agents and Artificial Intelligence part_num 2 page_num 857-864 publisher place Setúbal name SciTePress year 2019 editor name1 Rocha name2 A. editor name1 Steels name2 L. editor name1 van den Herik name2 J.
keyword	exploitation
keyword	exploration
keyword	adaptive systems
keyword	Bayesian estimation
keyword	fully probabilistic design
keyword	Markov decision process
author (primary)	ARLID cav_un_auth*0101124 name1 Kárný name2 Miroslav institution UTIA-B full_dept (cz) Adaptivní systémy full_dept (eng) Department of Adaptive Systems department (cz) AS department (eng) AS full_dept Department of Adaptive Systems fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
author	ARLID cav_un_auth*0333671 name1 Hůla name2 František institution UTIA-B full_dept (cz) Adaptivní systémy full_dept Department of Adaptive Systems department (cz) AS department AS full_dept Department of Adaptive Systems country CZ fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
source	url http://library.utia.cas.cz/separaty/2019/AS/hula-0503817.pdf
cas_special	project ARLID cav_un_auth*0331019 project_id GA16-09848S agency GA ČR country CZ project ARLID cav_un_auth*0362986 project_id GA18-15970S agency GA ČR country CZ abstract (eng) Adaptive decision making learns an environment model serving a design of a decision policy. The policy-generated actions influence both the acquired reward and the future knowledge. The optimal policy properly balances exploitation with exploration. The inherent dimensionality curse of decision making under incomplete knowledge prevents the realisation of the optimal design. This has stimulated repetitive attempts to reach this balance at least approximately. Usually, either: (a) the exploitative reward is enriched by a part reflecting the exploration quality and a feasible approximate certainty-equivalent design is made, or (b) an explorative random noise is added to the purely exploitative actions. This paper avoids the inauspicious (a) and improves (b) by employing the non-standard fully probabilistic design (FPD) of decision policies, which naturally generates random actions. Monte-Carlo experiments confirm its achieved quality. The quality stems from methodological contributions, which include: (i) an improvement of the relation between FPD and standard Markov decision processes, (ii) a design of an adaptive tuning of an FPD-parameter. The latter also suits for the tuning of the temperature in both simulated annealing and Boltzmann’s machine. action ARLID cav_un_auth*0374223 name International Conference on Agents and Artificial Intelligence dates 20190219 mrcbC20-s 20190221 place Praha country CZ RIV BC FORD0 10000 FORD1 10200 FORD2 10201 reportyear 2020 num_of_auth 2 mrcbC52 4 A sml 4as 20241106135724.9 presentation_type PO inst_support RVO:67985556 permalink http://hdl.handle.net/11104/0295689 confidential S contract name Content to publish and copyright transfer date 20190117 arlyear 2019 mrcbTft \nSoubory v repozitáři: hula-0503817-copyright.pdf mrcbU14 85064837601 SCOPUS mrcbU24 PUBMED mrcbU34 WOS mrcbU63 cav_un_epca*0503816 Proceedings of the 11th International Conference on Agents and Artificial Intelligence 2 SciTePress 2019 Setúbal 857 864 978-989-758-350-6 mrcbU67 340 Rocha A. mrcbU67 340 Steels L. mrcbU67 340 van den Herik J.