Matches in SemOpenAlex for { <https://semopenalex.org/work/W3213203216> ?p ?o ?g. }
- W3213203216 abstract "Neural bandits have enabled practitioners to operate efficiently on problems with non-linear reward functions. While in general contextual bandits commonly utilize Gaussian process (GP) predictive distributions for decision making, the most successful neural variants use only the last layer parameters in the derivation. Research on neural kernels (NK) has recently established a correspondence between deep networks and GPs that take into account all the parameters of a NN and can be trained more efficiently than most Bayesian NNs. We propose to directly apply NK-induced distributions to guide an upper confidence bound or Thompson sampling-based policy. We show that NK bandits achieve state-of-the-art performance on highly non-linear structured data. Furthermore, we analyze practical considerations such as training frequency and model partitioning. We believe our work will help better understand the impact of utilizing NKs in applied settings." @default.
- W3213203216 created "2021-11-22" @default.
- W3213203216 creator A5019604149 @default.
- W3213203216 creator A5034466911 @default.
- W3213203216 creator A5044546262 @default.
- W3213203216 date "2021-11-05" @default.
- W3213203216 modified "2023-09-27" @default.
- W3213203216 title "An Empirical Study of Neural Kernel Bandits" @default.
- W3213203216 cites W1487320471 @default.
- W3213203216 cites W1746819321 @default.
- W3213203216 cites W2039522160 @default.
- W3213203216 cites W2112420033 @default.
- W3213203216 cites W2166253248 @default.
- W3213203216 cites W2167608136 @default.
- W3213203216 cites W2168405694 @default.
- W3213203216 cites W2605172573 @default.
- W3213203216 cites W2766678531 @default.
- W3213203216 cites W2807588596 @default.
- W3213203216 cites W2809090039 @default.
- W3213203216 cites W2884810024 @default.
- W3213203216 cites W2942052807 @default.
- W3213203216 cites W2949915801 @default.
- W3213203216 cites W2950238385 @default.
- W3213203216 cites W2962939986 @default.
- W3213203216 cites W2964330179 @default.
- W3213203216 cites W2970217468 @default.
- W3213203216 cites W2993805018 @default.
- W3213203216 cites W2994872659 @default.
- W3213203216 cites W3015612994 @default.
- W3213203216 cites W3040219874 @default.
- W3213203216 cites W3042158842 @default.
- W3213203216 cites W3091650971 @default.
- W3213203216 cites W3109399063 @default.
- W3213203216 cites W3124229194 @default.
- W3213203216 cites W3128147955 @default.
- W3213203216 hasPublicationYear "2021" @default.
- W3213203216 type Work @default.
- W3213203216 sameAs 3213203216 @default.
- W3213203216 citedByCount "0" @default.
- W3213203216 crossrefType "posted-content" @default.
- W3213203216 hasAuthorship W3213203216A5019604149 @default.
- W3213203216 hasAuthorship W3213203216A5034466911 @default.
- W3213203216 hasAuthorship W3213203216A5044546262 @default.
- W3213203216 hasConcept C107673813 @default.
- W3213203216 hasConcept C111919701 @default.
- W3213203216 hasConcept C114614502 @default.
- W3213203216 hasConcept C119857082 @default.
- W3213203216 hasConcept C121332964 @default.
- W3213203216 hasConcept C154945302 @default.
- W3213203216 hasConcept C163716315 @default.
- W3213203216 hasConcept C2984842247 @default.
- W3213203216 hasConcept C33923547 @default.
- W3213203216 hasConcept C41008148 @default.
- W3213203216 hasConcept C50644808 @default.
- W3213203216 hasConcept C61326573 @default.
- W3213203216 hasConcept C62520636 @default.
- W3213203216 hasConcept C73602740 @default.
- W3213203216 hasConcept C74193536 @default.
- W3213203216 hasConcept C98045186 @default.
- W3213203216 hasConceptScore W3213203216C107673813 @default.
- W3213203216 hasConceptScore W3213203216C111919701 @default.
- W3213203216 hasConceptScore W3213203216C114614502 @default.
- W3213203216 hasConceptScore W3213203216C119857082 @default.
- W3213203216 hasConceptScore W3213203216C121332964 @default.
- W3213203216 hasConceptScore W3213203216C154945302 @default.
- W3213203216 hasConceptScore W3213203216C163716315 @default.
- W3213203216 hasConceptScore W3213203216C2984842247 @default.
- W3213203216 hasConceptScore W3213203216C33923547 @default.
- W3213203216 hasConceptScore W3213203216C41008148 @default.
- W3213203216 hasConceptScore W3213203216C50644808 @default.
- W3213203216 hasConceptScore W3213203216C61326573 @default.
- W3213203216 hasConceptScore W3213203216C62520636 @default.
- W3213203216 hasConceptScore W3213203216C73602740 @default.
- W3213203216 hasConceptScore W3213203216C74193536 @default.
- W3213203216 hasConceptScore W3213203216C98045186 @default.
- W3213203216 hasLocation W32132032161 @default.
- W3213203216 hasOpenAccess W3213203216 @default.
- W3213203216 hasPrimaryLocation W32132032161 @default.
- W3213203216 hasRelatedWork W2787512446 @default.
- W3213203216 hasRelatedWork W2804602357 @default.
- W3213203216 hasRelatedWork W2891630298 @default.
- W3213203216 hasRelatedWork W2898077722 @default.
- W3213203216 hasRelatedWork W2903099987 @default.
- W3213203216 hasRelatedWork W2913573237 @default.
- W3213203216 hasRelatedWork W2936952966 @default.
- W3213203216 hasRelatedWork W2950297419 @default.
- W3213203216 hasRelatedWork W2962750597 @default.
- W3213203216 hasRelatedWork W2970337468 @default.
- W3213203216 hasRelatedWork W3000569652 @default.
- W3213203216 hasRelatedWork W3014266848 @default.
- W3213203216 hasRelatedWork W3023570597 @default.
- W3213203216 hasRelatedWork W3082462885 @default.
- W3213203216 hasRelatedWork W3098708496 @default.
- W3213203216 hasRelatedWork W3112667379 @default.
- W3213203216 hasRelatedWork W3166771275 @default.
- W3213203216 hasRelatedWork W3170763958 @default.
- W3213203216 hasRelatedWork W4226468399 @default.
- W3213203216 hasRelatedWork W3096979466 @default.
- W3213203216 isParatext "false" @default.
- W3213203216 isRetracted "false" @default.