Matches in SemOpenAlex for { <https://semopenalex.org/work/W4294382023> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4294382023 abstract "In this paper we present a technique to train neural network models on small amounts of data. Current methods for training neural networks on small amounts of rich data typically rely on strategies such as fine-tuning a pre-trained neural network or the use of domain-specific hand-engineered features. Here we take the approach of treating network layers, or entire networks, as modules and combine pre-trained modules with untrained modules, to learn the shift in distributions between data sets. The central impact of using a modular approach comes from adding new representations to a network, as opposed to replacing representations via fine-tuning. Using this technique, we are able surpass results using standard fine-tuning transfer learning approaches, and we are also able to significantly increase performance over such approaches when using smaller amounts of data." @default.
- W4294382023 created "2022-09-03" @default.
- W4294382023 creator A5021941722 @default.
- W4294382023 creator A5023448303 @default.
- W4294382023 creator A5057070782 @default.
- W4294382023 creator A5079758945 @default.
- W4294382023 creator A5084366345 @default.
- W4294382023 date "2016-11-05" @default.
- W4294382023 modified "2023-09-27" @default.
- W4294382023 title "Beyond Fine Tuning: A Modular Approach to Learning on Small Data" @default.
- W4294382023 doi "https://doi.org/10.48550/arxiv.1611.01714" @default.
- W4294382023 hasPublicationYear "2016" @default.
- W4294382023 type Work @default.
- W4294382023 citedByCount "0" @default.
- W4294382023 crossrefType "posted-content" @default.
- W4294382023 hasAuthorship W4294382023A5021941722 @default.
- W4294382023 hasAuthorship W4294382023A5023448303 @default.
- W4294382023 hasAuthorship W4294382023A5057070782 @default.
- W4294382023 hasAuthorship W4294382023A5079758945 @default.
- W4294382023 hasAuthorship W4294382023A5084366345 @default.
- W4294382023 hasBestOaLocation W42943820231 @default.
- W4294382023 hasConcept C101468663 @default.
- W4294382023 hasConcept C111919701 @default.
- W4294382023 hasConcept C119857082 @default.
- W4294382023 hasConcept C121332964 @default.
- W4294382023 hasConcept C134306372 @default.
- W4294382023 hasConcept C150899416 @default.
- W4294382023 hasConcept C154945302 @default.
- W4294382023 hasConcept C157524613 @default.
- W4294382023 hasConcept C2779280203 @default.
- W4294382023 hasConcept C2984842247 @default.
- W4294382023 hasConcept C33923547 @default.
- W4294382023 hasConcept C36503486 @default.
- W4294382023 hasConcept C41008148 @default.
- W4294382023 hasConcept C50644808 @default.
- W4294382023 hasConcept C51632099 @default.
- W4294382023 hasConcept C62520636 @default.
- W4294382023 hasConceptScore W4294382023C101468663 @default.
- W4294382023 hasConceptScore W4294382023C111919701 @default.
- W4294382023 hasConceptScore W4294382023C119857082 @default.
- W4294382023 hasConceptScore W4294382023C121332964 @default.
- W4294382023 hasConceptScore W4294382023C134306372 @default.
- W4294382023 hasConceptScore W4294382023C150899416 @default.
- W4294382023 hasConceptScore W4294382023C154945302 @default.
- W4294382023 hasConceptScore W4294382023C157524613 @default.
- W4294382023 hasConceptScore W4294382023C2779280203 @default.
- W4294382023 hasConceptScore W4294382023C2984842247 @default.
- W4294382023 hasConceptScore W4294382023C33923547 @default.
- W4294382023 hasConceptScore W4294382023C36503486 @default.
- W4294382023 hasConceptScore W4294382023C41008148 @default.
- W4294382023 hasConceptScore W4294382023C50644808 @default.
- W4294382023 hasConceptScore W4294382023C51632099 @default.
- W4294382023 hasConceptScore W4294382023C62520636 @default.
- W4294382023 hasLocation W42943820231 @default.
- W4294382023 hasOpenAccess W4294382023 @default.
- W4294382023 hasPrimaryLocation W42943820231 @default.
- W4294382023 hasRelatedWork W2554354235 @default.
- W4294382023 hasRelatedWork W2760234924 @default.
- W4294382023 hasRelatedWork W2786459654 @default.
- W4294382023 hasRelatedWork W2966207284 @default.
- W4294382023 hasRelatedWork W3041434171 @default.
- W4294382023 hasRelatedWork W3099765033 @default.
- W4294382023 hasRelatedWork W3176808018 @default.
- W4294382023 hasRelatedWork W3207282234 @default.
- W4294382023 hasRelatedWork W4283785712 @default.
- W4294382023 hasRelatedWork W4292230744 @default.
- W4294382023 isParatext "false" @default.
- W4294382023 isRetracted "false" @default.
- W4294382023 workType "article" @default.