Matches in SemOpenAlex for { <https://semopenalex.org/work/W3040480737> ?p ?o ?g. }
Showing items 1 to 66 of
66
with 100 items per page.
- W3040480737 abstract "We study a structured variant of the multi-armed bandit problem specified by a set of Bernoulli distributions $ nu != !(nu_{a,b})_{a in mathcal{A}, b in mathcal{B}}$ with means $(mu_{a,b})_{a in mathcal{A}, b in mathcal{B}}!in![0,1]^{mathcal{A}timesmathcal{B}}$ and by a given weight matrix $omega!=! (omega_{b,b'})_{b,b' in mathcal{B}}$, where $ mathcal{A}$ is a finite set of arms and $ mathcal{B} $ is a finite set of users. The weight matrix $omega$ is such that for any two users $b,b'!in!mathcal{B}, text{max}_{ainmathcal{A}}|mu_{a,b} !-! mu_{a,b'}| !leq! omega_{b,b'} $. This formulation is flexible enough to capture various situations, from highly-structured scenarios ($omega!in!{0,1}^{mathcal{B}timesmathcal{B}}$) to fully unstructured setups ($omega!equiv! 1$).We consider two scenarios depending on whether the learner chooses only the actions to sample rewards from or both users and actions. We first derive problem-dependent lower bounds on the regret for this generic graph-structure that involves a structure dependent linear programming problem. Second, we adapt to this setting the Indexed Minimum Empirical Divergence (IMED) algorithm introduced by Honda and Takemura (2015), and introduce the IMED-GS$^star$ algorithm. Interestingly, IMED-GS$^star$ does not require computing the solution of the linear programming problem more than about $log(T)$ times after $T$ steps, while being provably asymptotically optimal. Also, unlike existing bandit strategies designed for other popular structures, IMED-GS$^star$ does not resort to an explicit forced exploration scheme and only makes use of local counts of empirical events. We finally provide numerical illustration of our results that confirm the performance of IMED-GS$^star$." @default.
- W3040480737 created "2020-07-10" @default.
- W3040480737 creator A5037297959 @default.
- W3040480737 creator A5048970129 @default.
- W3040480737 creator A5055126334 @default.
- W3040480737 date "2020-07-08" @default.
- W3040480737 modified "2023-10-14" @default.
- W3040480737 title "Optimal Strategies for Graph-Structured Bandits" @default.
- W3040480737 cites W1839697241 @default.
- W3040480737 cites W1973885534 @default.
- W3040480737 cites W1998376807 @default.
- W3040480737 cites W1998498767 @default.
- W3040480737 cites W2009551863 @default.
- W3040480737 cites W2013886500 @default.
- W3040480737 cites W2039522160 @default.
- W3040480737 cites W2119738618 @default.
- W3040480737 cites W2129036575 @default.
- W3040480737 cites W2131958277 @default.
- W3040480737 cites W2147398813 @default.
- W3040480737 cites W2165856466 @default.
- W3040480737 cites W2274664974 @default.
- W3040480737 cites W2295214655 @default.
- W3040480737 cites W2740620148 @default.
- W3040480737 cites W2785395147 @default.
- W3040480737 cites W2951590424 @default.
- W3040480737 cites W2951665052 @default.
- W3040480737 cites W2963465244 @default.
- W3040480737 cites W2963850079 @default.
- W3040480737 cites W2964000506 @default.
- W3040480737 cites W3100329718 @default.
- W3040480737 cites W3124229194 @default.
- W3040480737 hasPublicationYear "2020" @default.
- W3040480737 type Work @default.
- W3040480737 sameAs 3040480737 @default.
- W3040480737 citedByCount "0" @default.
- W3040480737 crossrefType "posted-content" @default.
- W3040480737 hasAuthorship W3040480737A5037297959 @default.
- W3040480737 hasAuthorship W3040480737A5048970129 @default.
- W3040480737 hasAuthorship W3040480737A5055126334 @default.
- W3040480737 hasBestOaLocation W30404807371 @default.
- W3040480737 hasConcept C132525143 @default.
- W3040480737 hasConcept C41008148 @default.
- W3040480737 hasConcept C80444323 @default.
- W3040480737 hasConceptScore W3040480737C132525143 @default.
- W3040480737 hasConceptScore W3040480737C41008148 @default.
- W3040480737 hasConceptScore W3040480737C80444323 @default.
- W3040480737 hasLocation W30404807371 @default.
- W3040480737 hasLocation W30404807372 @default.
- W3040480737 hasLocation W30404807373 @default.
- W3040480737 hasLocation W30404807374 @default.
- W3040480737 hasOpenAccess W3040480737 @default.
- W3040480737 hasPrimaryLocation W30404807371 @default.
- W3040480737 hasRelatedWork W1596801655 @default.
- W3040480737 hasRelatedWork W2130043461 @default.
- W3040480737 hasRelatedWork W2350741829 @default.
- W3040480737 hasRelatedWork W2358668433 @default.
- W3040480737 hasRelatedWork W2376932109 @default.
- W3040480737 hasRelatedWork W2382290278 @default.
- W3040480737 hasRelatedWork W2390279801 @default.
- W3040480737 hasRelatedWork W2748952813 @default.
- W3040480737 hasRelatedWork W2899084033 @default.
- W3040480737 hasRelatedWork W2530322880 @default.
- W3040480737 isParatext "false" @default.
- W3040480737 isRetracted "false" @default.
- W3040480737 magId "3040480737" @default.
- W3040480737 workType "article" @default.