Matches in SemOpenAlex for { <https://semopenalex.org/work/W4281800886> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4281800886 abstract "Recent works on neural contextual bandits have achieved compelling performances due to their ability to leverage the strong representation power of neural networks (NNs) for reward prediction. Many applications of contextual bandits involve multiple agents who collaborate without sharing raw observations, thus giving rise to the setting of federated contextual bandits. Existing works on federated contextual bandits rely on linear or kernelized bandits, which may fall short when modeling complex real-world reward functions. So, this paper introduces the federated neural-upper confidence bound (FN-UCB) algorithm. To better exploit the federated setting, FN-UCB adopts a weighted combination of two UCBs: $text{UCB}^{a}$ allows every agent to additionally use the observations from the other agents to accelerate exploration (without sharing raw observations), while $text{UCB}^{b}$ uses an NN with aggregated parameters for reward prediction in a similar way to federated averaging for supervised learning. Notably, the weight between the two UCBs required by our theoretical analysis is amenable to an interesting interpretation, which emphasizes $text{UCB}^{a}$ initially for accelerated exploration and relies more on $text{UCB}^{b}$ later after enough observations have been collected to train the NNs for accurate reward prediction (i.e., reliable exploitation). We prove sub-linear upper bounds on both the cumulative regret and the number of communication rounds of FN-UCB, and empirically demonstrate its competitive performance." @default.
- W4281800886 created "2022-06-13" @default.
- W4281800886 creator A5000956644 @default.
- W4281800886 creator A5018942019 @default.
- W4281800886 creator A5030304400 @default.
- W4281800886 creator A5045116715 @default.
- W4281800886 creator A5061691532 @default.
- W4281800886 creator A5064034052 @default.
- W4281800886 date "2022-05-27" @default.
- W4281800886 modified "2023-09-29" @default.
- W4281800886 title "Federated Neural Bandits" @default.
- W4281800886 doi "https://doi.org/10.48550/arxiv.2205.14309" @default.
- W4281800886 hasPublicationYear "2022" @default.
- W4281800886 type Work @default.
- W4281800886 citedByCount "0" @default.
- W4281800886 crossrefType "posted-content" @default.
- W4281800886 hasAuthorship W4281800886A5000956644 @default.
- W4281800886 hasAuthorship W4281800886A5018942019 @default.
- W4281800886 hasAuthorship W4281800886A5030304400 @default.
- W4281800886 hasAuthorship W4281800886A5045116715 @default.
- W4281800886 hasAuthorship W4281800886A5061691532 @default.
- W4281800886 hasAuthorship W4281800886A5064034052 @default.
- W4281800886 hasBestOaLocation W42818008861 @default.
- W4281800886 hasConcept C119857082 @default.
- W4281800886 hasConcept C153083717 @default.
- W4281800886 hasConcept C154945302 @default.
- W4281800886 hasConcept C165696696 @default.
- W4281800886 hasConcept C17744445 @default.
- W4281800886 hasConcept C199539241 @default.
- W4281800886 hasConcept C2776359362 @default.
- W4281800886 hasConcept C2984842247 @default.
- W4281800886 hasConcept C38652104 @default.
- W4281800886 hasConcept C41008148 @default.
- W4281800886 hasConcept C50644808 @default.
- W4281800886 hasConcept C50817715 @default.
- W4281800886 hasConcept C94625758 @default.
- W4281800886 hasConceptScore W4281800886C119857082 @default.
- W4281800886 hasConceptScore W4281800886C153083717 @default.
- W4281800886 hasConceptScore W4281800886C154945302 @default.
- W4281800886 hasConceptScore W4281800886C165696696 @default.
- W4281800886 hasConceptScore W4281800886C17744445 @default.
- W4281800886 hasConceptScore W4281800886C199539241 @default.
- W4281800886 hasConceptScore W4281800886C2776359362 @default.
- W4281800886 hasConceptScore W4281800886C2984842247 @default.
- W4281800886 hasConceptScore W4281800886C38652104 @default.
- W4281800886 hasConceptScore W4281800886C41008148 @default.
- W4281800886 hasConceptScore W4281800886C50644808 @default.
- W4281800886 hasConceptScore W4281800886C50817715 @default.
- W4281800886 hasConceptScore W4281800886C94625758 @default.
- W4281800886 hasLocation W42818008861 @default.
- W4281800886 hasOpenAccess W4281800886 @default.
- W4281800886 hasPrimaryLocation W42818008861 @default.
- W4281800886 hasRelatedWork W2130423968 @default.
- W4281800886 hasRelatedWork W2791246720 @default.
- W4281800886 hasRelatedWork W2936207217 @default.
- W4281800886 hasRelatedWork W2948262272 @default.
- W4281800886 hasRelatedWork W4226328666 @default.
- W4281800886 hasRelatedWork W4285428843 @default.
- W4281800886 hasRelatedWork W4307320276 @default.
- W4281800886 hasRelatedWork W4320165504 @default.
- W4281800886 hasRelatedWork W4322008322 @default.
- W4281800886 hasRelatedWork W1629725936 @default.
- W4281800886 isParatext "false" @default.
- W4281800886 isRetracted "false" @default.
- W4281800886 workType "article" @default.