Matches in SemOpenAlex for { <https://semopenalex.org/work/W2137329805> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2137329805 endingPage "2302" @default.
- W2137329805 startingPage "2294" @default.
- W2137329805 abstract "Most standard algorithms for prediction with expert advice depend on a parameter called the learning rate. This learning rate needs to be large enough to fit the data well, but small enough to prevent overfitting. For the exponential weights algorithm, a sequence of prior work has established theoretical guarantees for higher and higher data-dependent tunings of the learning rate, which allow for increasingly aggressive learning. But in practice such theoretical tunings often still perform worse (as measured by their regret) than ad hoc tuning with an even higher learning rate. To close the gap between theory and practice we introduce an approach to learn the learning rate. Up to a factor that is at most (poly)logarithmic in the number of experts and the inverse of the learning rate, our method performs as well as if we would know the empirically best learning rate from a large range that includes both conservative small values and values that are much higher than those for which formal guarantees were previously available. Our method employs a grid of learning rates, yet runs in linear time regardless of the size of the grid." @default.
- W2137329805 created "2016-06-24" @default.
- W2137329805 creator A5013618078 @default.
- W2137329805 creator A5063805934 @default.
- W2137329805 creator A5079050279 @default.
- W2137329805 date "2014-12-08" @default.
- W2137329805 modified "2023-09-24" @default.
- W2137329805 title "Learning the Learning Rate for Prediction with Expert Advice" @default.
- W2137329805 cites W1570963478 @default.
- W2137329805 cites W1808384435 @default.
- W2137329805 cites W186151502 @default.
- W2137329805 cites W1988790447 @default.
- W2137329805 cites W2005951230 @default.
- W2137329805 cites W2055639053 @default.
- W2137329805 cites W2093825590 @default.
- W2137329805 cites W2099111195 @default.
- W2137329805 cites W2149107760 @default.
- W2137329805 cites W2295562228 @default.
- W2137329805 cites W3149261725 @default.
- W2137329805 cites W2139716865 @default.
- W2137329805 hasPublicationYear "2014" @default.
- W2137329805 type Work @default.
- W2137329805 sameAs 2137329805 @default.
- W2137329805 citedByCount "6" @default.
- W2137329805 countsByYear W21373298052015 @default.
- W2137329805 countsByYear W21373298052017 @default.
- W2137329805 countsByYear W21373298052019 @default.
- W2137329805 countsByYear W21373298052020 @default.
- W2137329805 countsByYear W21373298052021 @default.
- W2137329805 crossrefType "proceedings-article" @default.
- W2137329805 hasAuthorship W2137329805A5013618078 @default.
- W2137329805 hasAuthorship W2137329805A5063805934 @default.
- W2137329805 hasAuthorship W2137329805A5079050279 @default.
- W2137329805 hasConcept C119857082 @default.
- W2137329805 hasConcept C134306372 @default.
- W2137329805 hasConcept C154945302 @default.
- W2137329805 hasConcept C187691185 @default.
- W2137329805 hasConcept C199360897 @default.
- W2137329805 hasConcept C22019652 @default.
- W2137329805 hasConcept C24138899 @default.
- W2137329805 hasConcept C2524010 @default.
- W2137329805 hasConcept C2779955035 @default.
- W2137329805 hasConcept C33923547 @default.
- W2137329805 hasConcept C39927690 @default.
- W2137329805 hasConcept C41008148 @default.
- W2137329805 hasConcept C50644808 @default.
- W2137329805 hasConcept C50817715 @default.
- W2137329805 hasConcept C77967617 @default.
- W2137329805 hasConceptScore W2137329805C119857082 @default.
- W2137329805 hasConceptScore W2137329805C134306372 @default.
- W2137329805 hasConceptScore W2137329805C154945302 @default.
- W2137329805 hasConceptScore W2137329805C187691185 @default.
- W2137329805 hasConceptScore W2137329805C199360897 @default.
- W2137329805 hasConceptScore W2137329805C22019652 @default.
- W2137329805 hasConceptScore W2137329805C24138899 @default.
- W2137329805 hasConceptScore W2137329805C2524010 @default.
- W2137329805 hasConceptScore W2137329805C2779955035 @default.
- W2137329805 hasConceptScore W2137329805C33923547 @default.
- W2137329805 hasConceptScore W2137329805C39927690 @default.
- W2137329805 hasConceptScore W2137329805C41008148 @default.
- W2137329805 hasConceptScore W2137329805C50644808 @default.
- W2137329805 hasConceptScore W2137329805C50817715 @default.
- W2137329805 hasConceptScore W2137329805C77967617 @default.
- W2137329805 hasLocation W21373298051 @default.
- W2137329805 hasOpenAccess W2137329805 @default.
- W2137329805 hasPrimaryLocation W21373298051 @default.
- W2137329805 hasRelatedWork W1483524195 @default.
- W2137329805 hasRelatedWork W1511670714 @default.
- W2137329805 hasRelatedWork W1551462465 @default.
- W2137329805 hasRelatedWork W1596504373 @default.
- W2137329805 hasRelatedWork W2068816944 @default.
- W2137329805 hasRelatedWork W2466644325 @default.
- W2137329805 hasRelatedWork W2736081685 @default.
- W2137329805 hasRelatedWork W2886249230 @default.
- W2137329805 hasRelatedWork W2894985519 @default.
- W2137329805 hasRelatedWork W2965483395 @default.
- W2137329805 hasRelatedWork W2997649681 @default.
- W2137329805 hasRelatedWork W3022865886 @default.
- W2137329805 hasRelatedWork W3026534141 @default.
- W2137329805 hasRelatedWork W3035422508 @default.
- W2137329805 hasRelatedWork W3046876458 @default.
- W2137329805 hasRelatedWork W3124872564 @default.
- W2137329805 hasRelatedWork W3201809877 @default.
- W2137329805 hasRelatedWork W3208307628 @default.
- W2137329805 hasRelatedWork W819164570 @default.
- W2137329805 hasRelatedWork W940094962 @default.
- W2137329805 hasVolume "27" @default.
- W2137329805 isParatext "false" @default.
- W2137329805 isRetracted "false" @default.
- W2137329805 magId "2137329805" @default.
- W2137329805 workType "article" @default.