Matches in SemOpenAlex for { <https://semopenalex.org/work/W2974920270> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W2974920270 abstract "We consider a stochastic linear bandit model in which the available actions correspond to arbitrary context vectors whose associated rewards follow a non-stationary linear regression model. In this setting, the unknown regression parameter is allowed to vary in time. To address this problem, we propose D-LinUCB, a novel optimistic algorithm based on discounted linear regression, where exponential weights are used to smoothly forget the past. This involves studying the deviations of the sequential weighted least-squares estimator under generic assumptions. As a by-product, we obtain novel deviation results that can be used beyond non-stationary environments. We provide theoretical guarantees on the behavior of D-LinUCB in both slowly-varying and abruptly-changing environments. We obtain an upper bound on the dynamic regret that is of order d^{2/3} B_T^{1/3}T^{2/3}, where B_T is a measure of non-stationarity (d and T being, respectively, dimension and horizon). This rate is known to be optimal. We also illustrate the empirical performance of D-LinUCB and compare it with recently proposed alternatives in simulated environments." @default.
- W2974920270 created "2019-09-26" @default.
- W2974920270 creator A5022571869 @default.
- W2974920270 creator A5049572169 @default.
- W2974920270 creator A5088066921 @default.
- W2974920270 date "2019-12-08" @default.
- W2974920270 modified "2023-09-25" @default.
- W2974920270 title "Weighted Linear Bandits for Non-Stationary Environments" @default.
- W2974920270 cites W1521230616 @default.
- W2974920270 cites W157259654 @default.
- W2974920270 cites W1628969036 @default.
- W2974920270 cites W2034284198 @default.
- W2974920270 cites W2069129115 @default.
- W2974920270 cites W2108114251 @default.
- W2974920270 cites W2112420033 @default.
- W2974920270 cites W2119738618 @default.
- W2974920270 cites W2149660380 @default.
- W2974920270 cites W2161571887 @default.
- W2974920270 cites W2736446958 @default.
- W2974920270 cites W2742123006 @default.
- W2974920270 cites W2742393003 @default.
- W2974920270 cites W2750611126 @default.
- W2974920270 cites W2786555653 @default.
- W2974920270 cites W2787284768 @default.
- W2974920270 cites W2787969715 @default.
- W2974920270 cites W2895748566 @default.
- W2974920270 cites W2902847006 @default.
- W2974920270 cites W2914791924 @default.
- W2974920270 cites W2917261185 @default.
- W2974920270 cites W2920202034 @default.
- W2974920270 cites W2950240626 @default.
- W2974920270 cites W2964133879 @default.
- W2974920270 cites W3112895936 @default.
- W2974920270 cites W3121159109 @default.
- W2974920270 cites W3123661679 @default.
- W2974920270 cites W3125579698 @default.
- W2974920270 hasPublicationYear "2019" @default.
- W2974920270 type Work @default.
- W2974920270 sameAs 2974920270 @default.
- W2974920270 citedByCount "10" @default.
- W2974920270 countsByYear W29749202702019 @default.
- W2974920270 countsByYear W29749202702020 @default.
- W2974920270 countsByYear W29749202702021 @default.
- W2974920270 crossrefType "proceedings-article" @default.
- W2974920270 hasAuthorship W2974920270A5022571869 @default.
- W2974920270 hasAuthorship W2974920270A5049572169 @default.
- W2974920270 hasAuthorship W2974920270A5088066921 @default.
- W2974920270 hasBestOaLocation W29749202701 @default.
- W2974920270 hasConcept C154945302 @default.
- W2974920270 hasConcept C41008148 @default.
- W2974920270 hasConceptScore W2974920270C154945302 @default.
- W2974920270 hasConceptScore W2974920270C41008148 @default.
- W2974920270 hasLocation W29749202701 @default.
- W2974920270 hasLocation W29749202702 @default.
- W2974920270 hasLocation W29749202703 @default.
- W2974920270 hasLocation W29749202704 @default.
- W2974920270 hasOpenAccess W2974920270 @default.
- W2974920270 hasPrimaryLocation W29749202701 @default.
- W2974920270 hasRelatedWork W2093578348 @default.
- W2974920270 hasRelatedWork W2096946506 @default.
- W2974920270 hasRelatedWork W2358668433 @default.
- W2974920270 hasRelatedWork W2376932109 @default.
- W2974920270 hasRelatedWork W2382290278 @default.
- W2974920270 hasRelatedWork W2390279801 @default.
- W2974920270 hasRelatedWork W2748952813 @default.
- W2974920270 hasRelatedWork W2766271392 @default.
- W2974920270 hasRelatedWork W2899084033 @default.
- W2974920270 hasRelatedWork W3107474891 @default.
- W2974920270 isParatext "false" @default.
- W2974920270 isRetracted "false" @default.
- W2974920270 magId "2974920270" @default.
- W2974920270 workType "article" @default.