Matches in SemOpenAlex for { <https://semopenalex.org/work/W1528131421> ?p ?o ?g. }
- W1528131421 abstract "The empirically successful Thompson Sampling algorithm for stochastic bandits has drawn much interest in understanding its theoretical properties. One important benefit of the algorithm is that it allows domain knowledge to be conveniently encoded as a prior distribution to balance exploration and exploitation more effectively. While it is generally believed that the algorithm's regret is low (high) when the prior is good (bad), little is known about the exact dependence. In this paper, we fully characterize the algorithm's worst-case dependence of regret on the choice of prior, focusing on a special yet representative case. These results also provide insights into the general sensitivity of the algorithm to the choice of priors. In particular, with $p$ being the prior probability mass of the true reward-generating model, we prove $O(sqrt{T/p})$ and $O(sqrt{(1-p)T})$ regret upper bounds for the bad- and good-prior cases, respectively, as well as emph{matching} lower bounds. Our proofs rely on the discovery of a fundamental property of Thompson Sampling and make heavy use of martingale theory, both of which appear novel in the literature, to the best of our knowledge." @default.
- W1528131421 created "2016-06-24" @default.
- W1528131421 creator A5030089269 @default.
- W1528131421 creator A5054850777 @default.
- W1528131421 date "2015-06-10" @default.
- W1528131421 modified "2023-09-26" @default.
- W1528131421 title "On the Prior Sensitivity of Thompson Sampling" @default.
- W1528131421 cites W148122774 @default.
- W1528131421 cites W1578424623 @default.
- W1528131421 cites W1603462117 @default.
- W1528131421 cites W1618543586 @default.
- W1528131421 cites W1676840945 @default.
- W1528131421 cites W1850547517 @default.
- W1528131421 cites W1911551976 @default.
- W1528131421 cites W2039522160 @default.
- W1528131421 cites W2049934117 @default.
- W1528131421 cites W2098258765 @default.
- W1528131421 cites W2108738385 @default.
- W1528131421 cites W2112925200 @default.
- W1528131421 cites W2119738618 @default.
- W1528131421 cites W2141645258 @default.
- W1528131421 cites W2149721706 @default.
- W1528131421 cites W2158125716 @default.
- W1528131421 cites W2162979096 @default.
- W1528131421 cites W2166253248 @default.
- W1528131421 cites W2951802169 @default.
- W1528131421 cites W2953138456 @default.
- W1528131421 cites W2962682763 @default.
- W1528131421 cites W2962901934 @default.
- W1528131421 cites W2964143446 @default.
- W1528131421 cites W611528516 @default.
- W1528131421 cites W2098339418 @default.
- W1528131421 doi "https://doi.org/10.48550/arxiv.1506.03378" @default.
- W1528131421 hasPublicationYear "2015" @default.
- W1528131421 type Work @default.
- W1528131421 sameAs 1528131421 @default.
- W1528131421 citedByCount "7" @default.
- W1528131421 countsByYear W15281314212015 @default.
- W1528131421 countsByYear W15281314212016 @default.
- W1528131421 countsByYear W15281314212019 @default.
- W1528131421 crossrefType "posted-content" @default.
- W1528131421 hasAuthorship W1528131421A5030089269 @default.
- W1528131421 hasAuthorship W1528131421A5054850777 @default.
- W1528131421 hasBestOaLocation W15281314211 @default.
- W1528131421 hasConcept C105795698 @default.
- W1528131421 hasConcept C106131492 @default.
- W1528131421 hasConcept C107673813 @default.
- W1528131421 hasConcept C108710211 @default.
- W1528131421 hasConcept C11413529 @default.
- W1528131421 hasConcept C114614502 @default.
- W1528131421 hasConcept C118615104 @default.
- W1528131421 hasConcept C127413603 @default.
- W1528131421 hasConcept C140779682 @default.
- W1528131421 hasConcept C144237770 @default.
- W1528131421 hasConcept C165064840 @default.
- W1528131421 hasConcept C177769412 @default.
- W1528131421 hasConcept C21200559 @default.
- W1528131421 hasConcept C24326235 @default.
- W1528131421 hasConcept C2524010 @default.
- W1528131421 hasConcept C28826006 @default.
- W1528131421 hasConcept C31972630 @default.
- W1528131421 hasConcept C33923547 @default.
- W1528131421 hasConcept C41008148 @default.
- W1528131421 hasConcept C48406656 @default.
- W1528131421 hasConcept C50817715 @default.
- W1528131421 hasConcept C73602740 @default.
- W1528131421 hasConceptScore W1528131421C105795698 @default.
- W1528131421 hasConceptScore W1528131421C106131492 @default.
- W1528131421 hasConceptScore W1528131421C107673813 @default.
- W1528131421 hasConceptScore W1528131421C108710211 @default.
- W1528131421 hasConceptScore W1528131421C11413529 @default.
- W1528131421 hasConceptScore W1528131421C114614502 @default.
- W1528131421 hasConceptScore W1528131421C118615104 @default.
- W1528131421 hasConceptScore W1528131421C127413603 @default.
- W1528131421 hasConceptScore W1528131421C140779682 @default.
- W1528131421 hasConceptScore W1528131421C144237770 @default.
- W1528131421 hasConceptScore W1528131421C165064840 @default.
- W1528131421 hasConceptScore W1528131421C177769412 @default.
- W1528131421 hasConceptScore W1528131421C21200559 @default.
- W1528131421 hasConceptScore W1528131421C24326235 @default.
- W1528131421 hasConceptScore W1528131421C2524010 @default.
- W1528131421 hasConceptScore W1528131421C28826006 @default.
- W1528131421 hasConceptScore W1528131421C31972630 @default.
- W1528131421 hasConceptScore W1528131421C33923547 @default.
- W1528131421 hasConceptScore W1528131421C41008148 @default.
- W1528131421 hasConceptScore W1528131421C48406656 @default.
- W1528131421 hasConceptScore W1528131421C50817715 @default.
- W1528131421 hasConceptScore W1528131421C73602740 @default.
- W1528131421 hasLocation W15281314211 @default.
- W1528131421 hasOpenAccess W1528131421 @default.
- W1528131421 hasPrimaryLocation W15281314211 @default.
- W1528131421 hasRelatedWork W1528131421 @default.
- W1528131421 hasRelatedWork W2116720812 @default.
- W1528131421 hasRelatedWork W2136101149 @default.
- W1528131421 hasRelatedWork W2951802169 @default.
- W1528131421 hasRelatedWork W3002095816 @default.
- W1528131421 hasRelatedWork W3108185025 @default.
- W1528131421 hasRelatedWork W3181522023 @default.
- W1528131421 hasRelatedWork W3196245787 @default.
- W1528131421 hasRelatedWork W4287079429 @default.