FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3120202350> ?p ?o ?g. }

• W3120202350 abstract "We study how representation learning can improve the efficiency of bandit problems. We study the setting where we play T linear bandits with dimension d concurrently, and these T bandit tasks share a common k(≪d) dimensional linear representation. For the finite-action setting, we present a new algorithm which achieves O~(TkN+dkNT) regret, where N is the number of rounds we play for each bandit. When T is sufficiently large, our algorithm significantly outperforms the naive algorithm (playing T bandits independently) that achieves O~(TdN) regret. We also provide an Ω(TkN+dkNT) regret lower bound, showing that our algorithm is minimax-optimal up to poly-logarithmic factors. Furthermore, we extend our algorithm to the infinite-action setting and obtain a corresponding regret bound which demonstrates the benefit of representation learning in certain regimes. We also present experiments on synthetic and real-world data to illustrate our theoretical findings and demonstrate the effectiveness of our proposed algorithms." @default.
• W3120202350 created "2021-01-18" @default.
• W3120202350 creator A5033061754 @default.
• W3120202350 creator A5056673247 @default.
• W3120202350 creator A5059740024 @default.
• W3120202350 creator A5062378128 @default.
• W3120202350 date "2021-05-03" @default.
• W3120202350 modified "2023-09-24" @default.
• W3120202350 title "Impact of Representation Learning in Linear Bandits" @default.
• W3120202350 cites W1487320471 @default.
• W3120202350 cites W2036043322 @default.
• W3120202350 cites W2061753713 @default.
• W3120202350 cites W2097381042 @default.
• W3120202350 cites W2108114251 @default.
• W3120202350 cites W2112420033 @default.
• W3120202350 cites W2116772053 @default.
• W3120202350 cites W2119738618 @default.
• W3120202350 cites W2122861691 @default.
• W3120202350 cites W2130903752 @default.
• W3120202350 cites W2142502798 @default.
• W3120202350 cites W2156267734 @default.
• W3120202350 cites W2163922914 @default.
• W3120202350 cites W2174786457 @default.
• W3120202350 cites W2237537322 @default.
• W3120202350 cites W2342750929 @default.
• W3120202350 cites W2592998600 @default.
• W3120202350 cites W2740053211 @default.
• W3120202350 cites W2787248994 @default.
• W3120202350 cites W2891076394 @default.
• W3120202350 cites W2912606190 @default.
• W3120202350 cites W2913340405 @default.
• W3120202350 cites W2923966100 @default.
• W3120202350 cites W2926246612 @default.
• W3120202350 cites W2943101255 @default.
• W3120202350 cites W2944804155 @default.
• W3120202350 cites W2951585248 @default.
• W3120202350 cites W2963070905 @default.
• W3120202350 cites W2963199420 @default.
• W3120202350 cites W2963649256 @default.
• W3120202350 cites W2964206659 @default.
• W3120202350 cites W2964211955 @default.
• W3120202350 cites W2965112429 @default.
• W3120202350 cites W2971037079 @default.
• W3120202350 cites W2972989214 @default.
• W3120202350 cites W2974885182 @default.
• W3120202350 cites W2980113592 @default.
• W3120202350 cites W2995481444 @default.
• W3120202350 cites W3006977545 @default.
• W3120202350 cites W3007684729 @default.
• W3120202350 cites W3017342111 @default.
• W3120202350 cites W3035180992 @default.
• W3120202350 cites W3158917068 @default.
• W3120202350 cites W3170045214 @default.
• W3120202350 cites W3211898143 @default.
• W3120202350 cites W50486269 @default.
• W3120202350 cites W2970619409 @default.
• W3120202350 hasPublicationYear "2021" @default.
• W3120202350 type Work @default.
• W3120202350 sameAs 3120202350 @default.
• W3120202350 citedByCount "4" @default.
• W3120202350 countsByYear W31202023502021 @default.
• W3120202350 crossrefType "proceedings-article" @default.
• W3120202350 hasAuthorship W3120202350A5033061754 @default.
• W3120202350 hasAuthorship W3120202350A5056673247 @default.
• W3120202350 hasAuthorship W3120202350A5059740024 @default.
• W3120202350 hasAuthorship W3120202350A5062378128 @default.
• W3120202350 hasConcept C11413529 @default.
• W3120202350 hasConcept C114614502 @default.
• W3120202350 hasConcept C119857082 @default.
• W3120202350 hasConcept C121332964 @default.
• W3120202350 hasConcept C126255220 @default.
• W3120202350 hasConcept C134306372 @default.
• W3120202350 hasConcept C149728462 @default.
• W3120202350 hasConcept C154945302 @default.
• W3120202350 hasConcept C17744445 @default.
• W3120202350 hasConcept C199539241 @default.
• W3120202350 hasConcept C2776359362 @default.
• W3120202350 hasConcept C2780791683 @default.
• W3120202350 hasConcept C33676613 @default.
• W3120202350 hasConcept C33923547 @default.
• W3120202350 hasConcept C39927690 @default.
• W3120202350 hasConcept C41008148 @default.
• W3120202350 hasConcept C50817715 @default.
• W3120202350 hasConcept C62520636 @default.
• W3120202350 hasConcept C77553402 @default.
• W3120202350 hasConcept C94625758 @default.
• W3120202350 hasConceptScore W3120202350C11413529 @default.
• W3120202350 hasConceptScore W3120202350C114614502 @default.
• W3120202350 hasConceptScore W3120202350C119857082 @default.
• W3120202350 hasConceptScore W3120202350C121332964 @default.
• W3120202350 hasConceptScore W3120202350C126255220 @default.
• W3120202350 hasConceptScore W3120202350C134306372 @default.
• W3120202350 hasConceptScore W3120202350C149728462 @default.
• W3120202350 hasConceptScore W3120202350C154945302 @default.
• W3120202350 hasConceptScore W3120202350C17744445 @default.
• W3120202350 hasConceptScore W3120202350C199539241 @default.
• W3120202350 hasConceptScore W3120202350C2776359362 @default.
• W3120202350 hasConceptScore W3120202350C2780791683 @default.
• W3120202350 hasConceptScore W3120202350C33676613 @default.
• W3120202350 hasConceptScore W3120202350C33923547 @default.

• next