Matches in SemOpenAlex for { <https://semopenalex.org/work/W3120876295> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W3120876295 abstract "This work presents a naive algorithm for parameter transfer between different architectures with a computationally cheap injection technique (which does not require data). The primary objective is to speed up the training of neural networks from scratch. It was found in this study that transferring knowledge from any architecture was superior to Kaiming and Xavier for initialization. In conclusion, the method presented is found to converge faster, which makes it a drop-in replacement for classical methods. The method involves: 1) matching: the layers of the pre-trained model with the targeted model; 2) injection: the tensor is transformed into a desired shape. This work provides a comparison of similarity between the current SOTA architectures (ImageNet), by utilising TLI (Transfer Learning by Injection) score." @default.
- W3120876295 created "2021-01-18" @default.
- W3120876295 creator A5055856998 @default.
- W3120876295 date "2021-01-07" @default.
- W3120876295 modified "2023-09-27" @default.
- W3120876295 title "Transfer Learning Between Different Architectures Via Weights Injection." @default.
- W3120876295 cites W1677182931 @default.
- W3120876295 cites W2178031510 @default.
- W3120876295 cites W2946948417 @default.
- W3120876295 cites W3111133646 @default.
- W3120876295 hasPublicationYear "2021" @default.
- W3120876295 type Work @default.
- W3120876295 sameAs 3120876295 @default.
- W3120876295 citedByCount "0" @default.
- W3120876295 crossrefType "posted-content" @default.
- W3120876295 hasAuthorship W3120876295A5055856998 @default.
- W3120876295 hasConcept C103278499 @default.
- W3120876295 hasConcept C105795698 @default.
- W3120876295 hasConcept C111919701 @default.
- W3120876295 hasConcept C114466953 @default.
- W3120876295 hasConcept C115961682 @default.
- W3120876295 hasConcept C119857082 @default.
- W3120876295 hasConcept C123657996 @default.
- W3120876295 hasConcept C142362112 @default.
- W3120876295 hasConcept C150899416 @default.
- W3120876295 hasConcept C153349607 @default.
- W3120876295 hasConcept C154945302 @default.
- W3120876295 hasConcept C155281189 @default.
- W3120876295 hasConcept C165064840 @default.
- W3120876295 hasConcept C199360897 @default.
- W3120876295 hasConcept C202444582 @default.
- W3120876295 hasConcept C2781235140 @default.
- W3120876295 hasConcept C33923547 @default.
- W3120876295 hasConcept C41008148 @default.
- W3120876295 hasConcept C50644808 @default.
- W3120876295 hasConceptScore W3120876295C103278499 @default.
- W3120876295 hasConceptScore W3120876295C105795698 @default.
- W3120876295 hasConceptScore W3120876295C111919701 @default.
- W3120876295 hasConceptScore W3120876295C114466953 @default.
- W3120876295 hasConceptScore W3120876295C115961682 @default.
- W3120876295 hasConceptScore W3120876295C119857082 @default.
- W3120876295 hasConceptScore W3120876295C123657996 @default.
- W3120876295 hasConceptScore W3120876295C142362112 @default.
- W3120876295 hasConceptScore W3120876295C150899416 @default.
- W3120876295 hasConceptScore W3120876295C153349607 @default.
- W3120876295 hasConceptScore W3120876295C154945302 @default.
- W3120876295 hasConceptScore W3120876295C155281189 @default.
- W3120876295 hasConceptScore W3120876295C165064840 @default.
- W3120876295 hasConceptScore W3120876295C199360897 @default.
- W3120876295 hasConceptScore W3120876295C202444582 @default.
- W3120876295 hasConceptScore W3120876295C2781235140 @default.
- W3120876295 hasConceptScore W3120876295C33923547 @default.
- W3120876295 hasConceptScore W3120876295C41008148 @default.
- W3120876295 hasConceptScore W3120876295C50644808 @default.
- W3120876295 hasLocation W31208762951 @default.
- W3120876295 hasOpenAccess W3120876295 @default.
- W3120876295 hasPrimaryLocation W31208762951 @default.
- W3120876295 hasRelatedWork W2135719924 @default.
- W3120876295 hasRelatedWork W2160082767 @default.
- W3120876295 hasRelatedWork W2618541429 @default.
- W3120876295 hasRelatedWork W2743024835 @default.
- W3120876295 hasRelatedWork W2766207311 @default.
- W3120876295 hasRelatedWork W2891874923 @default.
- W3120876295 hasRelatedWork W2898200114 @default.
- W3120876295 hasRelatedWork W2903655494 @default.
- W3120876295 hasRelatedWork W2912543964 @default.
- W3120876295 hasRelatedWork W2940669070 @default.
- W3120876295 hasRelatedWork W2951388117 @default.
- W3120876295 hasRelatedWork W2979556024 @default.
- W3120876295 hasRelatedWork W3024070930 @default.
- W3120876295 hasRelatedWork W3091889904 @default.
- W3120876295 hasRelatedWork W3101272433 @default.
- W3120876295 hasRelatedWork W3124157622 @default.
- W3120876295 hasRelatedWork W3130602232 @default.
- W3120876295 hasRelatedWork W3133313904 @default.
- W3120876295 hasRelatedWork W3177188747 @default.
- W3120876295 hasRelatedWork W79084438 @default.
- W3120876295 isParatext "false" @default.
- W3120876295 isRetracted "false" @default.
- W3120876295 magId "3120876295" @default.
- W3120876295 workType "article" @default.