Matches in SemOpenAlex for { <https://semopenalex.org/work/W3203522710> ?p ?o ?g. }
- W3203522710 abstract "We consider a contextual bandit problem with a combinatorial action set and time-varying base arm availability. At the beginning of each round, the agent observes the set of available base arms and their contexts and then selects an action that is a feasible subset of the set of available base arms to maximize its cumulative reward in the long run. We assume that the mean outcomes of base arms are samples from a Gaussian Process indexed by the context set ${cal X}$, and the expected reward is Lipschitz continuous in expected base arm outcomes. For this setup, we propose an algorithm called Optimistic Combinatorial Learning and Optimization with Kernel Upper Confidence Bounds (O'CLOK-UCB) and prove that it incurs $tilde{O}(Ksqrt{Toverline{gamma}_{T}} )$ regret with high probability, where $overline{gamma}_{T}$ is the maximum information gain associated with the set of base arm contexts that appeared in the first $T$ rounds and $K$ is the maximum cardinality of any feasible action over all rounds. To dramatically speed up the algorithm, we also propose a variant of O'CLOK-UCB that uses sparse GPs. Finally, we experimentally show that both algorithms exploit inter-base arm outcome correlation and vastly outperform the previous state-of-the-art UCB-based algorithms in realistic setups." @default.
- W3203522710 created "2021-10-11" @default.
- W3203522710 creator A5031100822 @default.
- W3203522710 creator A5053015746 @default.
- W3203522710 creator A5056816244 @default.
- W3203522710 date "2021-10-05" @default.
- W3203522710 modified "2023-09-27" @default.
- W3203522710 title "Contextual Combinatorial Volatile Bandits via Gaussian Processes" @default.
- W3203522710 cites W103689294 @default.
- W3203522710 cites W137285897 @default.
- W3203522710 cites W1487320471 @default.
- W3203522710 cites W1597810135 @default.
- W3203522710 cites W1680189815 @default.
- W3203522710 cites W1900560890 @default.
- W3203522710 cites W1917528016 @default.
- W3203522710 cites W1955698431 @default.
- W3203522710 cites W1968143987 @default.
- W3203522710 cites W2008098735 @default.
- W3203522710 cites W2049934117 @default.
- W3203522710 cites W2071702404 @default.
- W3203522710 cites W2077902449 @default.
- W3203522710 cites W2093562354 @default.
- W3203522710 cites W2120090487 @default.
- W3203522710 cites W2132801025 @default.
- W3203522710 cites W2166566250 @default.
- W3203522710 cites W2168405694 @default.
- W3203522710 cites W2183950117 @default.
- W3203522710 cites W2219888463 @default.
- W3203522710 cites W2222512263 @default.
- W3203522710 cites W2237158255 @default.
- W3203522710 cites W2400213106 @default.
- W3203522710 cites W2401264332 @default.
- W3203522710 cites W2519411794 @default.
- W3203522710 cites W2540189295 @default.
- W3203522710 cites W2914156981 @default.
- W3203522710 cites W2962839911 @default.
- W3203522710 cites W2964007796 @default.
- W3203522710 cites W2990738158 @default.
- W3203522710 cites W3086778302 @default.
- W3203522710 cites W3102150134 @default.
- W3203522710 cites W35251828 @default.
- W3203522710 cites W2890020788 @default.
- W3203522710 hasPublicationYear "2021" @default.
- W3203522710 type Work @default.
- W3203522710 sameAs 3203522710 @default.
- W3203522710 citedByCount "1" @default.
- W3203522710 countsByYear W32035227102021 @default.
- W3203522710 crossrefType "posted-content" @default.
- W3203522710 hasAuthorship W3203522710A5031100822 @default.
- W3203522710 hasAuthorship W3203522710A5053015746 @default.
- W3203522710 hasAuthorship W3203522710A5056816244 @default.
- W3203522710 hasConcept C11413529 @default.
- W3203522710 hasConcept C114614502 @default.
- W3203522710 hasConcept C118615104 @default.
- W3203522710 hasConcept C119857082 @default.
- W3203522710 hasConcept C121332964 @default.
- W3203522710 hasConcept C124101348 @default.
- W3203522710 hasConcept C126255220 @default.
- W3203522710 hasConcept C134306372 @default.
- W3203522710 hasConcept C14036430 @default.
- W3203522710 hasConcept C151730666 @default.
- W3203522710 hasConcept C163716315 @default.
- W3203522710 hasConcept C165696696 @default.
- W3203522710 hasConcept C177264268 @default.
- W3203522710 hasConcept C199360897 @default.
- W3203522710 hasConcept C22324862 @default.
- W3203522710 hasConcept C2779343474 @default.
- W3203522710 hasConcept C2780791683 @default.
- W3203522710 hasConcept C33923547 @default.
- W3203522710 hasConcept C38652104 @default.
- W3203522710 hasConcept C41008148 @default.
- W3203522710 hasConcept C42058472 @default.
- W3203522710 hasConcept C50817715 @default.
- W3203522710 hasConcept C61326573 @default.
- W3203522710 hasConcept C62520636 @default.
- W3203522710 hasConcept C73602740 @default.
- W3203522710 hasConcept C74193536 @default.
- W3203522710 hasConcept C78458016 @default.
- W3203522710 hasConcept C86803240 @default.
- W3203522710 hasConcept C87117476 @default.
- W3203522710 hasConceptScore W3203522710C11413529 @default.
- W3203522710 hasConceptScore W3203522710C114614502 @default.
- W3203522710 hasConceptScore W3203522710C118615104 @default.
- W3203522710 hasConceptScore W3203522710C119857082 @default.
- W3203522710 hasConceptScore W3203522710C121332964 @default.
- W3203522710 hasConceptScore W3203522710C124101348 @default.
- W3203522710 hasConceptScore W3203522710C126255220 @default.
- W3203522710 hasConceptScore W3203522710C134306372 @default.
- W3203522710 hasConceptScore W3203522710C14036430 @default.
- W3203522710 hasConceptScore W3203522710C151730666 @default.
- W3203522710 hasConceptScore W3203522710C163716315 @default.
- W3203522710 hasConceptScore W3203522710C165696696 @default.
- W3203522710 hasConceptScore W3203522710C177264268 @default.
- W3203522710 hasConceptScore W3203522710C199360897 @default.
- W3203522710 hasConceptScore W3203522710C22324862 @default.
- W3203522710 hasConceptScore W3203522710C2779343474 @default.
- W3203522710 hasConceptScore W3203522710C2780791683 @default.
- W3203522710 hasConceptScore W3203522710C33923547 @default.
- W3203522710 hasConceptScore W3203522710C38652104 @default.
- W3203522710 hasConceptScore W3203522710C41008148 @default.