Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385682007> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4385682007 abstract "In this paper, we investigate transfer learning in partially observable contextual bandits, where agents have limited knowledge from other agents and partial information about hidden confounders. We first convert the problem to identifying or partially identifying causal effects between actions and rewards through optimization problems. To solve these optimization problems, we discretize the original functional constraints of unknown distributions into linear constraints, and sample compatible causal models via sequentially solving linear programmings to obtain causal bounds with the consideration of estimation error. Our sampling algorithms provide desirable convergence results for suitable sampling distributions. We then show how causal bounds can be applied to improving classical bandit algorithms and affect the regrets with respect to the size of action sets and function spaces. Notably, in the task with function approximation which allows us to handle general context distributions, our method improves the order dependence on function space size compared with previous literatures. We formally prove that our causally enhanced algorithms outperform classical bandit algorithms and achieve orders of magnitude faster convergence rates. Finally, we perform simulations that demonstrate the efficiency of our strategy compared to the current state-of-the-art methods. This research has the potential to enhance the performance of contextual bandit agents in real-world applications where data is scarce and costly to obtain." @default.
- W4385682007 created "2023-08-09" @default.
- W4385682007 creator A5004297494 @default.
- W4385682007 creator A5035377116 @default.
- W4385682007 date "2023-08-07" @default.
- W4385682007 modified "2023-09-27" @default.
- W4385682007 title "Provably Efficient Learning in Partially Observable Contextual Bandit" @default.
- W4385682007 doi "https://doi.org/10.48550/arxiv.2308.03572" @default.
- W4385682007 hasPublicationYear "2023" @default.
- W4385682007 type Work @default.
- W4385682007 citedByCount "0" @default.
- W4385682007 crossrefType "posted-content" @default.
- W4385682007 hasAuthorship W4385682007A5004297494 @default.
- W4385682007 hasAuthorship W4385682007A5035377116 @default.
- W4385682007 hasBestOaLocation W43856820071 @default.
- W4385682007 hasConcept C106131492 @default.
- W4385682007 hasConcept C107673813 @default.
- W4385682007 hasConcept C11413529 @default.
- W4385682007 hasConcept C121332964 @default.
- W4385682007 hasConcept C126255220 @default.
- W4385682007 hasConcept C134306372 @default.
- W4385682007 hasConcept C14036430 @default.
- W4385682007 hasConcept C140779682 @default.
- W4385682007 hasConcept C151730666 @default.
- W4385682007 hasConcept C154945302 @default.
- W4385682007 hasConcept C162324750 @default.
- W4385682007 hasConcept C2777303404 @default.
- W4385682007 hasConcept C2779343474 @default.
- W4385682007 hasConcept C31972630 @default.
- W4385682007 hasConcept C32848918 @default.
- W4385682007 hasConcept C33923547 @default.
- W4385682007 hasConcept C41008148 @default.
- W4385682007 hasConcept C50522688 @default.
- W4385682007 hasConcept C62520636 @default.
- W4385682007 hasConcept C73000952 @default.
- W4385682007 hasConcept C73602740 @default.
- W4385682007 hasConcept C78458016 @default.
- W4385682007 hasConcept C86803240 @default.
- W4385682007 hasConceptScore W4385682007C106131492 @default.
- W4385682007 hasConceptScore W4385682007C107673813 @default.
- W4385682007 hasConceptScore W4385682007C11413529 @default.
- W4385682007 hasConceptScore W4385682007C121332964 @default.
- W4385682007 hasConceptScore W4385682007C126255220 @default.
- W4385682007 hasConceptScore W4385682007C134306372 @default.
- W4385682007 hasConceptScore W4385682007C14036430 @default.
- W4385682007 hasConceptScore W4385682007C140779682 @default.
- W4385682007 hasConceptScore W4385682007C151730666 @default.
- W4385682007 hasConceptScore W4385682007C154945302 @default.
- W4385682007 hasConceptScore W4385682007C162324750 @default.
- W4385682007 hasConceptScore W4385682007C2777303404 @default.
- W4385682007 hasConceptScore W4385682007C2779343474 @default.
- W4385682007 hasConceptScore W4385682007C31972630 @default.
- W4385682007 hasConceptScore W4385682007C32848918 @default.
- W4385682007 hasConceptScore W4385682007C33923547 @default.
- W4385682007 hasConceptScore W4385682007C41008148 @default.
- W4385682007 hasConceptScore W4385682007C50522688 @default.
- W4385682007 hasConceptScore W4385682007C62520636 @default.
- W4385682007 hasConceptScore W4385682007C73000952 @default.
- W4385682007 hasConceptScore W4385682007C73602740 @default.
- W4385682007 hasConceptScore W4385682007C78458016 @default.
- W4385682007 hasConceptScore W4385682007C86803240 @default.
- W4385682007 hasLocation W43856820071 @default.
- W4385682007 hasOpenAccess W4385682007 @default.
- W4385682007 hasPrimaryLocation W43856820071 @default.
- W4385682007 hasRelatedWork W2040186499 @default.
- W4385682007 hasRelatedWork W2066091055 @default.
- W4385682007 hasRelatedWork W2083355284 @default.
- W4385682007 hasRelatedWork W2146591199 @default.
- W4385682007 hasRelatedWork W2163596130 @default.
- W4385682007 hasRelatedWork W2360394982 @default.
- W4385682007 hasRelatedWork W2952302283 @default.
- W4385682007 hasRelatedWork W4297726110 @default.
- W4385682007 hasRelatedWork W4377003522 @default.
- W4385682007 hasRelatedWork W85224956 @default.
- W4385682007 isParatext "false" @default.
- W4385682007 isRetracted "false" @default.
- W4385682007 workType "article" @default.