Matches in SemOpenAlex for { <https://semopenalex.org/work/W2157549058> ?p ?o ?g. }
- W2157549058 endingPage "642" @default.
- W2157549058 startingPage "631" @default.
- W2157549058 abstract "We propose a theoretical and computational framework for approximating the optimal policy in multi-armed bandit problems where the reward distributions are non-Gaussian. We first construct a probabilistic interpolation of the sequence of discrete-time rewards in the form of a continuous-time conditional Levy process. In the Gaussian setting, this approach allows an easy connection to Brownian motion and its convenient time-change properties. No such device is available for non-Gaussian rewards; however, we show how optimal stopping theory can be used to characterize the value of the optimal policy, using a free-boundary partial integro-differential equation, for exponential and Poisson rewards. We then solve this problem numerically to approximate the set of belief states possessing a given optimal index value, and provide illustrations showing that the solution behaves as expected." @default.
- W2157549058 created "2016-06-24" @default.
- W2157549058 creator A5027345703 @default.
- W2157549058 creator A5044367354 @default.
- W2157549058 date "2013-12-08" @default.
- W2157549058 modified "2023-10-16" @default.
- W2157549058 title "Optimal learning with non-Gaussian rewards" @default.
- W2157549058 cites W1492655370 @default.
- W2157549058 cites W1508957056 @default.
- W2157549058 cites W1513215787 @default.
- W2157549058 cites W1525688252 @default.
- W2157549058 cites W1539851785 @default.
- W2157549058 cites W1545370368 @default.
- W2157549058 cites W1546899016 @default.
- W2157549058 cites W171917236 @default.
- W2157549058 cites W1965288387 @default.
- W2157549058 cites W1973476981 @default.
- W2157549058 cites W1988360637 @default.
- W2157549058 cites W1995408121 @default.
- W2157549058 cites W1995442741 @default.
- W2157549058 cites W1998353334 @default.
- W2157549058 cites W2000080679 @default.
- W2157549058 cites W2006258746 @default.
- W2157549058 cites W2009543467 @default.
- W2157549058 cites W2009551863 @default.
- W2157549058 cites W2015685007 @default.
- W2157549058 cites W2017334003 @default.
- W2157549058 cites W2027338865 @default.
- W2157549058 cites W2036554718 @default.
- W2157549058 cites W2037728958 @default.
- W2157549058 cites W2040106856 @default.
- W2157549058 cites W2042332971 @default.
- W2157549058 cites W2042353373 @default.
- W2157549058 cites W2044825177 @default.
- W2157549058 cites W2045426407 @default.
- W2157549058 cites W2049687551 @default.
- W2157549058 cites W2057564132 @default.
- W2157549058 cites W2061367129 @default.
- W2157549058 cites W2063853624 @default.
- W2157549058 cites W2068420430 @default.
- W2157549058 cites W2069034916 @default.
- W2157549058 cites W2072302356 @default.
- W2157549058 cites W2072709311 @default.
- W2157549058 cites W2078011655 @default.
- W2157549058 cites W2082247722 @default.
- W2157549058 cites W2082343755 @default.
- W2157549058 cites W2087590712 @default.
- W2157549058 cites W2089425220 @default.
- W2157549058 cites W2100922150 @default.
- W2157549058 cites W2106554316 @default.
- W2157549058 cites W2110674418 @default.
- W2157549058 cites W2111787450 @default.
- W2157549058 cites W2111983268 @default.
- W2157549058 cites W2113425147 @default.
- W2157549058 cites W2118309135 @default.
- W2157549058 cites W2124832196 @default.
- W2157549058 cites W2127306098 @default.
- W2157549058 cites W2140571193 @default.
- W2157549058 cites W2166274157 @default.
- W2157549058 cites W2168405694 @default.
- W2157549058 cites W2185196136 @default.
- W2157549058 cites W2250110863 @default.
- W2157549058 cites W2317700292 @default.
- W2157549058 cites W2331686122 @default.
- W2157549058 cites W2499002200 @default.
- W2157549058 cites W2499226730 @default.
- W2157549058 cites W2797641211 @default.
- W2157549058 cites W3122796486 @default.
- W2157549058 cites W3136506851 @default.
- W2157549058 cites W604762380 @default.
- W2157549058 cites W615510159 @default.
- W2157549058 cites W620776879 @default.
- W2157549058 doi "https://doi.org/10.5555/2675983.2676066" @default.
- W2157549058 hasPublicationYear "2013" @default.
- W2157549058 type Work @default.
- W2157549058 sameAs 2157549058 @default.
- W2157549058 citedByCount "0" @default.
- W2157549058 crossrefType "proceedings-article" @default.
- W2157549058 hasAuthorship W2157549058A5027345703 @default.
- W2157549058 hasAuthorship W2157549058A5044367354 @default.
- W2157549058 hasConcept C104114177 @default.
- W2157549058 hasConcept C105795698 @default.
- W2157549058 hasConcept C112401455 @default.
- W2157549058 hasConcept C121332964 @default.
- W2157549058 hasConcept C126255220 @default.
- W2157549058 hasConcept C137800194 @default.
- W2157549058 hasConcept C14646407 @default.
- W2157549058 hasConcept C154945302 @default.
- W2157549058 hasConcept C163716315 @default.
- W2157549058 hasConcept C2778112365 @default.
- W2157549058 hasConcept C28826006 @default.
- W2157549058 hasConcept C33923547 @default.
- W2157549058 hasConcept C41008148 @default.
- W2157549058 hasConcept C49937458 @default.
- W2157549058 hasConcept C54355233 @default.
- W2157549058 hasConcept C61326573 @default.
- W2157549058 hasConcept C62520636 @default.
- W2157549058 hasConcept C86803240 @default.