Matches in SemOpenAlex for { <https://semopenalex.org/work/W1569268747> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W1569268747 abstract "Reinforcement learning methods are increasingly used to optimise dialogue policies from experience. Most current techniques are model-free: they directly estimate the utility of various actions, without explicit model of the interaction dynamics. In this paper, we investigate an alternative strategy grounded in model-based Bayesian reinforcement learning. Bayesian inference is used to maintain a posterior distribution over the model parameters, reflecting the model uncertainty. This parameter distribution is gradually refined as more data is collected and simultaneously used to plan the agent’s actions. Within this learning framework, we carried out experiments with two alternative formalisations of the transition model, one encoded with standard multinomial distributions, and one structured with probabilistic rules. We demonstrate the potential of our approach with empirical results on a user simulator constructed from Wizard-of-Oz data in a human–robot interaction scenario. The results illustrate in particular the benefits of capturing prior domain knowledge with high-level rules." @default.
- W1569268747 created "2016-06-24" @default.
- W1569268747 creator A5008574874 @default.
- W1569268747 date "2013-08-25" @default.
- W1569268747 modified "2023-09-23" @default.
- W1569268747 title "Model-based Bayesian reinforcement learning for dialogue management" @default.
- W1569268747 cites W1532688806 @default.
- W1569268747 cites W1587057191 @default.
- W1569268747 cites W1592751638 @default.
- W1569268747 cites W1757949796 @default.
- W1569268747 cites W190008395 @default.
- W1569268747 cites W1951650891 @default.
- W1569268747 cites W1965555277 @default.
- W1569268747 cites W1972247907 @default.
- W1569268747 cites W1975244201 @default.
- W1569268747 cites W1985992429 @default.
- W1569268747 cites W2015936967 @default.
- W1569268747 cites W2021151961 @default.
- W1569268747 cites W2037897789 @default.
- W1569268747 cites W2040123554 @default.
- W1569268747 cites W2083849869 @default.
- W1569268747 cites W2095977914 @default.
- W1569268747 cites W2099430963 @default.
- W1569268747 cites W2112476714 @default.
- W1569268747 cites W2115101920 @default.
- W1569268747 cites W2119015791 @default.
- W1569268747 cites W2121863487 @default.
- W1569268747 cites W2132997613 @default.
- W1569268747 cites W2144913588 @default.
- W1569268747 cites W2151904921 @default.
- W1569268747 cites W2153385324 @default.
- W1569268747 cites W2158640361 @default.
- W1569268747 cites W2160422981 @default.
- W1569268747 cites W2168839459 @default.
- W1569268747 cites W2171084228 @default.
- W1569268747 cites W2341171179 @default.
- W1569268747 cites W2396935534 @default.
- W1569268747 cites W2408200822 @default.
- W1569268747 cites W2438667436 @default.
- W1569268747 cites W2624335110 @default.
- W1569268747 cites W37568705 @default.
- W1569268747 doi "https://doi.org/10.21437/interspeech.2013-138" @default.
- W1569268747 hasPublicationYear "2013" @default.
- W1569268747 type Work @default.
- W1569268747 sameAs 1569268747 @default.
- W1569268747 citedByCount "7" @default.
- W1569268747 countsByYear W15692687472014 @default.
- W1569268747 countsByYear W15692687472015 @default.
- W1569268747 countsByYear W15692687472016 @default.
- W1569268747 countsByYear W15692687472021 @default.
- W1569268747 crossrefType "proceedings-article" @default.
- W1569268747 hasAuthorship W1569268747A5008574874 @default.
- W1569268747 hasBestOaLocation W15692687472 @default.
- W1569268747 hasConcept C107673813 @default.
- W1569268747 hasConcept C119857082 @default.
- W1569268747 hasConcept C154945302 @default.
- W1569268747 hasConcept C41008148 @default.
- W1569268747 hasConcept C97541855 @default.
- W1569268747 hasConceptScore W1569268747C107673813 @default.
- W1569268747 hasConceptScore W1569268747C119857082 @default.
- W1569268747 hasConceptScore W1569268747C154945302 @default.
- W1569268747 hasConceptScore W1569268747C41008148 @default.
- W1569268747 hasConceptScore W1569268747C97541855 @default.
- W1569268747 hasLocation W15692687471 @default.
- W1569268747 hasLocation W15692687472 @default.
- W1569268747 hasLocation W15692687473 @default.
- W1569268747 hasOpenAccess W1569268747 @default.
- W1569268747 hasPrimaryLocation W15692687471 @default.
- W1569268747 hasRelatedWork W1562959674 @default.
- W1569268747 hasRelatedWork W2140035747 @default.
- W1569268747 hasRelatedWork W2923653485 @default.
- W1569268747 hasRelatedWork W2952472710 @default.
- W1569268747 hasRelatedWork W2961085424 @default.
- W1569268747 hasRelatedWork W3022038857 @default.
- W1569268747 hasRelatedWork W3173482257 @default.
- W1569268747 hasRelatedWork W3209094908 @default.
- W1569268747 hasRelatedWork W4210912933 @default.
- W1569268747 hasRelatedWork W4255994452 @default.
- W1569268747 isParatext "false" @default.
- W1569268747 isRetracted "false" @default.
- W1569268747 magId "1569268747" @default.
- W1569268747 workType "article" @default.