Matches in SemOpenAlex for { <https://semopenalex.org/work/W3038554660> ?p ?o ?g. }
Showing items 1 to 66 of
66
with 100 items per page.
- W3038554660 abstract "We study reward maximisation in a wide class of structured stochastic multi-armed bandit problems, where the mean rewards of arms satisfy some given structural constraints, e.g. linear, unimodal, sparse, etc. Our aim is to develop methods that are flexible (in that they easily adapt to different structures), powerful (in that they perform well empirically and/or provably match instance-dependent lower bounds) and efficient in that the per-round computational burden is small. We develop asymptotically optimal algorithms from instance-dependent lower-bounds using iterative saddle-point solvers. Our approach generalises recent iterative methods for pure exploration to reward maximisation, where a major challenge arises from the estimation of the sub-optimality gaps and their reciprocals. Still we manage to achieve all the above desiderata. Notably, our technique avoids the computational cost of the full-blown saddle point oracle employed by previous work, while at the same time enabling finite-time regret bounds. Our experiments reveal that our method successfully leverages the structural assumptions, while its regret is at worst comparable to that of vanilla UCB." @default.
- W3038554660 created "2020-07-10" @default.
- W3038554660 creator A5013618078 @default.
- W3038554660 creator A5050128377 @default.
- W3038554660 creator A5074202131 @default.
- W3038554660 date "2020-07-12" @default.
- W3038554660 modified "2023-09-27" @default.
- W3038554660 title "Structure Adaptive Algorithms for Stochastic Bandits" @default.
- W3038554660 cites W2000080679 @default.
- W3038554660 cites W2009551863 @default.
- W3038554660 cites W2028404160 @default.
- W3038554660 cites W2039522160 @default.
- W3038554660 cites W2049934117 @default.
- W3038554660 cites W2119738618 @default.
- W3038554660 cites W2129036575 @default.
- W3038554660 cites W2147398813 @default.
- W3038554660 cites W2149107760 @default.
- W3038554660 cites W2168405694 @default.
- W3038554660 cites W2296319761 @default.
- W3038554660 cites W2903138945 @default.
- W3038554660 cites W2914156981 @default.
- W3038554660 cites W2951590424 @default.
- W3038554660 cites W2963275573 @default.
- W3038554660 cites W2963465244 @default.
- W3038554660 cites W2963486287 @default.
- W3038554660 cites W2963750583 @default.
- W3038554660 cites W2963850079 @default.
- W3038554660 cites W2963968798 @default.
- W3038554660 cites W2964007796 @default.
- W3038554660 cites W2970201409 @default.
- W3038554660 cites W2970242045 @default.
- W3038554660 cites W2970271647 @default.
- W3038554660 cites W50486269 @default.
- W3038554660 hasPublicationYear "2020" @default.
- W3038554660 type Work @default.
- W3038554660 sameAs 3038554660 @default.
- W3038554660 citedByCount "3" @default.
- W3038554660 countsByYear W30385546602020 @default.
- W3038554660 crossrefType "proceedings-article" @default.
- W3038554660 hasAuthorship W3038554660A5013618078 @default.
- W3038554660 hasAuthorship W3038554660A5050128377 @default.
- W3038554660 hasAuthorship W3038554660A5074202131 @default.
- W3038554660 hasBestOaLocation W30385546602 @default.
- W3038554660 hasConcept C11413529 @default.
- W3038554660 hasConcept C41008148 @default.
- W3038554660 hasConceptScore W3038554660C11413529 @default.
- W3038554660 hasConceptScore W3038554660C41008148 @default.
- W3038554660 hasLocation W30385546601 @default.
- W3038554660 hasLocation W30385546602 @default.
- W3038554660 hasLocation W30385546603 @default.
- W3038554660 hasOpenAccess W3038554660 @default.
- W3038554660 hasPrimaryLocation W30385546601 @default.
- W3038554660 hasRelatedWork W2051487156 @default.
- W3038554660 hasRelatedWork W2073681303 @default.
- W3038554660 hasRelatedWork W2317200988 @default.
- W3038554660 hasRelatedWork W2358668433 @default.
- W3038554660 hasRelatedWork W2376932109 @default.
- W3038554660 hasRelatedWork W2382290278 @default.
- W3038554660 hasRelatedWork W2386767533 @default.
- W3038554660 hasRelatedWork W2390279801 @default.
- W3038554660 hasRelatedWork W2748952813 @default.
- W3038554660 hasRelatedWork W2899084033 @default.
- W3038554660 isParatext "false" @default.
- W3038554660 isRetracted "false" @default.
- W3038554660 magId "3038554660" @default.
- W3038554660 workType "article" @default.