Matches in SemOpenAlex for { <https://semopenalex.org/work/W2998085019> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2998085019 endingPage "8001" @default.
- W2998085019 startingPage "7994" @default.
- W2998085019 abstract "We consider a strategic dialogue task, where the ability to infer the other agent's goal is critical to the success of the conversational agent. While this problem can be naturally formulated as Bayesian planning, it is known to be a very difficult problem due to its enormous search space consisting of all possible utterances. In this paper, we introduce an efficient Bayes-adaptive planning algorithm for goal-oriented dialogues, which combines RNN-based dialogue generation and MCTS-based Bayesian planning in a novel way, leading to robust decision-making under the uncertainty of the other agent's goal. We then introduce reinforcement learning for the dialogue agent that uses MCTS as a strong policy improvement operator, casting reinforcement learning as iterative alternation of planning and supervised-learning of self-generated dialogues. In the experiments, we demonstrate that our Bayes-adaptive dialogue planning agent significantly outperforms the state-of-the-art in a negotiation dialogue domain. We also show that reinforcement learning via MCTS further improves end-task performance without diverging from human language." @default.
- W2998085019 created "2020-01-10" @default.
- W2998085019 creator A5015283188 @default.
- W2998085019 creator A5031835179 @default.
- W2998085019 creator A5021442487 @default.
- W2998085019 date "2020-04-03" @default.
- W2998085019 modified "2023-09-27" @default.
- W2998085019 title "Bayes-Adaptive Monte-Carlo Planning and Learning for Goal-Oriented Dialogues" @default.
- W2998085019 cites W1500868819 @default.
- W2998085019 cites W1625390266 @default.
- W2998085019 cites W2075848246 @default.
- W2998085019 cites W2119717200 @default.
- W2998085019 cites W2126316555 @default.
- W2998085019 cites W2149586740 @default.
- W2998085019 cites W2157477959 @default.
- W2998085019 cites W2168405694 @default.
- W2998085019 cites W2464790259 @default.
- W2998085019 cites W2468710617 @default.
- W2998085019 cites W2766447205 @default.
- W2998085019 cites W2902907165 @default.
- W2998085019 cites W2915295540 @default.
- W2998085019 cites W2962852262 @default.
- W2998085019 cites W2962879001 @default.
- W2998085019 cites W2963134326 @default.
- W2998085019 cites W2963167310 @default.
- W2998085019 cites W2963330684 @default.
- W2998085019 cites W2963343509 @default.
- W2998085019 cites W2963730239 @default.
- W2998085019 cites W2963790827 @default.
- W2998085019 cites W2963797754 @default.
- W2998085019 cites W2964101860 @default.
- W2998085019 cites W2964268978 @default.
- W2998085019 cites W2964308564 @default.
- W2998085019 doi "https://doi.org/10.1609/aaai.v34i05.6308" @default.
- W2998085019 hasPublicationYear "2020" @default.
- W2998085019 type Work @default.
- W2998085019 sameAs 2998085019 @default.
- W2998085019 citedByCount "8" @default.
- W2998085019 countsByYear W29980850192020 @default.
- W2998085019 countsByYear W29980850192021 @default.
- W2998085019 crossrefType "journal-article" @default.
- W2998085019 hasAuthorship W2998085019A5015283188 @default.
- W2998085019 hasAuthorship W2998085019A5021442487 @default.
- W2998085019 hasAuthorship W2998085019A5031835179 @default.
- W2998085019 hasBestOaLocation W29980850191 @default.
- W2998085019 hasConcept C107673813 @default.
- W2998085019 hasConcept C119857082 @default.
- W2998085019 hasConcept C134306372 @default.
- W2998085019 hasConcept C154945302 @default.
- W2998085019 hasConcept C162324750 @default.
- W2998085019 hasConcept C17744445 @default.
- W2998085019 hasConcept C187736073 @default.
- W2998085019 hasConcept C199539241 @default.
- W2998085019 hasConcept C199776023 @default.
- W2998085019 hasConcept C207201462 @default.
- W2998085019 hasConcept C2780451532 @default.
- W2998085019 hasConcept C33923547 @default.
- W2998085019 hasConcept C36503486 @default.
- W2998085019 hasConcept C41008148 @default.
- W2998085019 hasConcept C97541855 @default.
- W2998085019 hasConceptScore W2998085019C107673813 @default.
- W2998085019 hasConceptScore W2998085019C119857082 @default.
- W2998085019 hasConceptScore W2998085019C134306372 @default.
- W2998085019 hasConceptScore W2998085019C154945302 @default.
- W2998085019 hasConceptScore W2998085019C162324750 @default.
- W2998085019 hasConceptScore W2998085019C17744445 @default.
- W2998085019 hasConceptScore W2998085019C187736073 @default.
- W2998085019 hasConceptScore W2998085019C199539241 @default.
- W2998085019 hasConceptScore W2998085019C199776023 @default.
- W2998085019 hasConceptScore W2998085019C207201462 @default.
- W2998085019 hasConceptScore W2998085019C2780451532 @default.
- W2998085019 hasConceptScore W2998085019C33923547 @default.
- W2998085019 hasConceptScore W2998085019C36503486 @default.
- W2998085019 hasConceptScore W2998085019C41008148 @default.
- W2998085019 hasConceptScore W2998085019C97541855 @default.
- W2998085019 hasIssue "05" @default.
- W2998085019 hasLocation W29980850191 @default.
- W2998085019 hasOpenAccess W2998085019 @default.
- W2998085019 hasPrimaryLocation W29980850191 @default.
- W2998085019 hasRelatedWork W1561788187 @default.
- W2998085019 hasRelatedWork W2046384965 @default.
- W2998085019 hasRelatedWork W2115714256 @default.
- W2998085019 hasRelatedWork W2140035747 @default.
- W2998085019 hasRelatedWork W2950680678 @default.
- W2998085019 hasRelatedWork W3004547119 @default.
- W2998085019 hasRelatedWork W3022038857 @default.
- W2998085019 hasRelatedWork W3034669169 @default.
- W2998085019 hasRelatedWork W4287870652 @default.
- W2998085019 hasRelatedWork W4302156121 @default.
- W2998085019 hasVolume "34" @default.
- W2998085019 isParatext "false" @default.
- W2998085019 isRetracted "false" @default.
- W2998085019 magId "2998085019" @default.
- W2998085019 workType "article" @default.