FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891146651> ?p ?o ?g. }

• W2891146651 endingPage "6855" @default.
• W2891146651 startingPage "6845" @default.
• W2891146651 abstract "Bayesian optimization (BO) is a model-based approach for gradient-free black-box function optimization, such as hyperparameter optimization. Typically, BO relies on conventional Gaussian process (GP) regression, whose algorithmic complexity is cubic in the number of evaluations. As a result, GP-based BO cannot leverage large numbers of past function evaluations, for example, to warm-start related BO runs. We propose a multi-task adaptive Bayesian linear regression model for transfer learning in BO, whose complexity is linear in the function evaluations: one Bayesian linear regression model is associated to each black-box function optimization problem (or task), while transfer learning is achieved by coupling the models through a shared deep neural net. Experiments show that the neural net learns a representation suitable for warm-starting the black-box optimization problems and that BO runs can be accelerated when the target black-box function (e.g., validation loss) is learned together with other related signals (e.g., training loss). The proposed method was found to be at least one order of magnitude faster that methods recently published in the literature." @default.
• W2891146651 created "2018-09-27" @default.
• W2891146651 creator A5005681375 @default.
• W2891146651 creator A5021107714 @default.
• W2891146651 creator A5033851429 @default.
• W2891146651 creator A5085696195 @default.
• W2891146651 date "2018-01-01" @default.
• W2891146651 modified "2023-09-24" @default.
• W2891146651 title "Scalable Hyperparameter Transfer Learning" @default.
• W2891146651 cites W122178443 @default.
• W2891146651 cites W137285897 @default.
• W2891146651 cites W1510052597 @default.
• W2891146651 cites W1512098439 @default.
• W2891146651 cites W1567512734 @default.
• W2891146651 cites W1571870753 @default.
• W2891146651 cites W1663203009 @default.
• W2891146651 cites W1701825639 @default.
• W2891146651 cites W1930835045 @default.
• W2891146651 cites W1959608418 @default.
• W2891146651 cites W2101709642 @default.
• W2891146651 cites W2111241577 @default.
• W2891146651 cites W2113145584 @default.
• W2891146651 cites W2119187866 @default.
• W2891146651 cites W2132862423 @default.
• W2891146651 cites W2137135057 @default.
• W2891146651 cites W2139479120 @default.
• W2891146651 cites W2144902422 @default.
• W2891146651 cites W2153635508 @default.
• W2891146651 cites W2182070128 @default.
• W2891146651 cites W2200000192 @default.
• W2891146651 cites W2296059279 @default.
• W2891146651 cites W2507221917 @default.
• W2891146651 cites W2556372419 @default.
• W2891146651 cites W2571670780 @default.
• W2891146651 cites W2622957538 @default.
• W2891146651 cites W2732547613 @default.
• W2891146651 cites W2734383650 @default.
• W2891146651 cites W2741261073 @default.
• W2891146651 cites W2765648398 @default.
• W2891146651 cites W2950103693 @default.
• W2891146651 cites W2963742654 @default.
• W2891146651 cites W2964121744 @default.
• W2891146651 cites W3140968660 @default.
• W2891146651 cites W76331760 @default.
• W2891146651 hasPublicationYear "2018" @default.
• W2891146651 type Work @default.
• W2891146651 sameAs 2891146651 @default.
• W2891146651 citedByCount "55" @default.
• W2891146651 countsByYear W28911466512018 @default.
• W2891146651 countsByYear W28911466512019 @default.
• W2891146651 countsByYear W28911466512020 @default.
• W2891146651 countsByYear W28911466512021 @default.
• W2891146651 countsByYear W28911466512022 @default.
• W2891146651 crossrefType "proceedings-article" @default.
• W2891146651 hasAuthorship W2891146651A5005681375 @default.
• W2891146651 hasAuthorship W2891146651A5021107714 @default.
• W2891146651 hasAuthorship W2891146651A5033851429 @default.
• W2891146651 hasAuthorship W2891146651A5085696195 @default.
• W2891146651 hasConcept C107673813 @default.
• W2891146651 hasConcept C108583219 @default.
• W2891146651 hasConcept C11413529 @default.
• W2891146651 hasConcept C119857082 @default.
• W2891146651 hasConcept C121332964 @default.
• W2891146651 hasConcept C150899416 @default.
• W2891146651 hasConcept C153083717 @default.
• W2891146651 hasConcept C154945302 @default.
• W2891146651 hasConcept C160234255 @default.
• W2891146651 hasConcept C163716315 @default.
• W2891146651 hasConcept C2778049539 @default.
• W2891146651 hasConcept C37903108 @default.
• W2891146651 hasConcept C41008148 @default.
• W2891146651 hasConcept C61326573 @default.
• W2891146651 hasConcept C62520636 @default.
• W2891146651 hasConcept C81692654 @default.
• W2891146651 hasConcept C8642999 @default.
• W2891146651 hasConcept C94966114 @default.
• W2891146651 hasConceptScore W2891146651C107673813 @default.
• W2891146651 hasConceptScore W2891146651C108583219 @default.
• W2891146651 hasConceptScore W2891146651C11413529 @default.
• W2891146651 hasConceptScore W2891146651C119857082 @default.
• W2891146651 hasConceptScore W2891146651C121332964 @default.
• W2891146651 hasConceptScore W2891146651C150899416 @default.
• W2891146651 hasConceptScore W2891146651C153083717 @default.
• W2891146651 hasConceptScore W2891146651C154945302 @default.
• W2891146651 hasConceptScore W2891146651C160234255 @default.
• W2891146651 hasConceptScore W2891146651C163716315 @default.
• W2891146651 hasConceptScore W2891146651C2778049539 @default.
• W2891146651 hasConceptScore W2891146651C37903108 @default.
• W2891146651 hasConceptScore W2891146651C41008148 @default.
• W2891146651 hasConceptScore W2891146651C61326573 @default.
• W2891146651 hasConceptScore W2891146651C62520636 @default.
• W2891146651 hasConceptScore W2891146651C81692654 @default.
• W2891146651 hasConceptScore W2891146651C8642999 @default.
• W2891146651 hasConceptScore W2891146651C94966114 @default.
• W2891146651 hasLocation W28911466511 @default.
• W2891146651 hasOpenAccess W2891146651 @default.
• W2891146651 hasPrimaryLocation W28911466511 @default.
• W2891146651 hasRelatedWork W122178443 @default.

• next