Počet záznamů: 1
Agent’s Feedback in Preference Elicitation
- 1.
SYSNO ASEP 0555371 Druh ASEP C - Konferenční příspěvek (mezinárodní konf.) Zařazení RIV D - Článek ve sborníku Název Agent’s Feedback in Preference Elicitation Tvůrce(i) Kárný, Miroslav (UTIA-B) RID, ORCID
Siváková, Tereza (UTIA-B) ORCIDCelkový počet autorů 2 Zdroj.dok. International Conference on Ubiquitous Computing and Communications and International Symposium on Cyberspace and Security (IUCC-CSS) 2021. - Piscataway : IEEE Computer Society, 2021 - ISBN 978-1-6654-6667-7 Rozsah stran s. 421-429 Poč.str. 9 s. Forma vydání Online - E Akce International Conference on Ubiquitous Computing and Communications 2021 (IUCC/CIT/DSCI/SmartCNS 2021) /20./ Datum konání 20.12.2021 - 22.12.2021 Místo konání London Země GB - Velká Británie Typ akce WRD Jazyk dok. eng - angličtina Země vyd. US - Spojené státy americké Klíč. slova Preference elicitation ; Adaptive agent ; Decision making ; Bayes’ rule Vědní obor RIV BC - Teorie a systémy řízení Obor OECD Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) CEP LTC18075 GA MŠMT - Ministerstvo školství, mládeže a tělovýchovy Institucionální podpora UTIA-B - RVO:67985556 UT WOS 000803071400058 EID SCOPUS 85127623905 DOI 10.1109/IUCC-CIT-DSCI-SmartCNS55181.2021.00073 Anotace A generic decision-making (DM) agent specifies its preferences partially. The studied prescriptiveDMtheory, called fully probabilistic design (FPD) of decision strategies, has recently addressed this obstacle in a new way. The found preference completion and quantification exploits that: IFPD models the closed DM loop and the agent’s preferences by joint probability densities (pds), Ithere is a preference-elicitation (PE) principle, which maps the agent’s model of the state transitions and its incompletely expressed wishes on an ideal pd quantifying them. The gained algorithmic uantification provides ambitious but potentially reachable DM aims. It suppresses demands on the agent selecting the preference-expressing inputs. The remaining PE options are: Ia parameter balancing exploration with exploitation, Ia fine specification of the ideal (desired) sets of states and actions, Irelative importance of these ideal sets. The current paper makes decisive steps towards a systematic and realistic choice of such inputs by solving a meta-DM task. The algorithmic “meta-agent” observes the user’s satisfaction, expressed by school-type marks, and tunes the free PE inputs to improve these marks. The solution requires a suitable formalisation of such a meta-task. This is done here. The proposed way copes with the danger of infinite regress and the imensionality curse. Non-trivial simulations illustrate the results. Pracoviště Ústav teorie informace a automatizace Kontakt Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Rok sběru 2023
Počet záznamů: 1