FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2947706148> ?p ?o ?g. }

• W2947706148 abstract "Transfer learning from natural image datasets, particularly ImageNet, using standard large models and corresponding pretrained weights has become a de-facto method for deep learning applications to medical imaging. However, there are fundamental differences in data sizes, features and task specifications between natural image classification and the target medical tasks, and there is little understanding of the effects of transfer. In this paper, we explore properties of transfer learning for medical imaging. A performance evaluation on two large scale medical imaging tasks shows that surprisingly, transfer offers little benefit to performance, and simple, lightweight models can perform comparably to ImageNet architectures. Investigating the learned representations and features, we find that some of the differences from transfer learning are due to the over-parametrization of standard models rather than sophisticated feature reuse. We isolate where useful feature reuse occurs, and outline the implications for more efficient model exploration. We also explore feature independent benefits of transfer arising from weight scalings." @default.
• W2947706148 created "2019-06-07" @default.
• W2947706148 creator A5016054994 @default.
• W2947706148 creator A5017529415 @default.
• W2947706148 creator A5027041171 @default.
• W2947706148 creator A5055710645 @default.
• W2947706148 date "2019-02-14" @default.
• W2947706148 modified "2023-10-01" @default.
• W2947706148 title "Transfusion: Understanding Transfer Learning for Medical Imaging" @default.
• W2947706148 cites W1915485278 @default.
• W2947706148 cites W2069454046 @default.
• W2947706148 cites W2097117768 @default.
• W2947706148 cites W2110091014 @default.
• W2947706148 cites W2149933564 @default.
• W2947706148 cites W2194775991 @default.
• W2947706148 cites W2342750929 @default.
• W2947706148 cites W2474421929 @default.
• W2947706148 cites W2510153535 @default.
• W2947706148 cites W2557738935 @default.
• W2947706148 cites W2581082771 @default.
• W2947706148 cites W2593345132 @default.
• W2947706148 cites W2767204723 @default.
• W2947706148 cites W2768095311 @default.
• W2947706148 cites W2770241596 @default.
• W2947706148 cites W2798512429 @default.
• W2947706148 cites W2804935296 @default.
• W2947706148 cites W2810697931 @default.
• W2947706148 cites W2886281300 @default.
• W2947706148 cites W2899402383 @default.
• W2947706148 cites W2899635607 @default.
• W2947706148 cites W2901026139 @default.
• W2947706148 cites W2901394229 @default.
• W2947706148 cites W2902617128 @default.
• W2947706148 cites W2913190747 @default.
• W2947706148 cites W2939788146 @default.
• W2947706148 cites W2949117887 @default.
• W2947706148 cites W2949558627 @default.
• W2947706148 cites W2950338985 @default.
• W2947706148 cites W2963337368 @default.
• W2947706148 cites W2963446712 @default.
• W2947706148 cites W2963466845 @default.
• W2947706148 cites W2963639714 @default.
• W2947706148 cites W2963759780 @default.
• W2947706148 cites W2964013315 @default.
• W2947706148 cites W2970316683 @default.
• W2947706148 cites W2971531230 @default.
• W2947706148 cites W3101156210 @default.
• W2947706148 cites W3118608800 @default.
• W2947706148 hasPublicationYear "2019" @default.
• W2947706148 type Work @default.
• W2947706148 sameAs 2947706148 @default.
• W2947706148 citedByCount "54" @default.
• W2947706148 countsByYear W29477061482019 @default.
• W2947706148 countsByYear W29477061482020 @default.
• W2947706148 countsByYear W29477061482021 @default.
• W2947706148 crossrefType "posted-content" @default.
• W2947706148 hasAuthorship W2947706148A5016054994 @default.
• W2947706148 hasAuthorship W2947706148A5017529415 @default.
• W2947706148 hasAuthorship W2947706148A5027041171 @default.
• W2947706148 hasAuthorship W2947706148A5055710645 @default.
• W2947706148 hasConcept C108583219 @default.
• W2947706148 hasConcept C111472728 @default.
• W2947706148 hasConcept C119857082 @default.
• W2947706148 hasConcept C121332964 @default.
• W2947706148 hasConcept C127413603 @default.
• W2947706148 hasConcept C138885662 @default.
• W2947706148 hasConcept C150899416 @default.
• W2947706148 hasConcept C154945302 @default.
• W2947706148 hasConcept C17744445 @default.
• W2947706148 hasConcept C199539241 @default.
• W2947706148 hasConcept C201995342 @default.
• W2947706148 hasConcept C202887219 @default.
• W2947706148 hasConcept C206588197 @default.
• W2947706148 hasConcept C2776401178 @default.
• W2947706148 hasConcept C2780451532 @default.
• W2947706148 hasConcept C2780586882 @default.
• W2947706148 hasConcept C2992317946 @default.
• W2947706148 hasConcept C31601959 @default.
• W2947706148 hasConcept C41008148 @default.
• W2947706148 hasConcept C41895202 @default.
• W2947706148 hasConcept C548081761 @default.
• W2947706148 hasConcept C62520636 @default.
• W2947706148 hasConcept C74902906 @default.
• W2947706148 hasConceptScore W2947706148C108583219 @default.
• W2947706148 hasConceptScore W2947706148C111472728 @default.
• W2947706148 hasConceptScore W2947706148C119857082 @default.
• W2947706148 hasConceptScore W2947706148C121332964 @default.
• W2947706148 hasConceptScore W2947706148C127413603 @default.
• W2947706148 hasConceptScore W2947706148C138885662 @default.
• W2947706148 hasConceptScore W2947706148C150899416 @default.
• W2947706148 hasConceptScore W2947706148C154945302 @default.
• W2947706148 hasConceptScore W2947706148C17744445 @default.
• W2947706148 hasConceptScore W2947706148C199539241 @default.
• W2947706148 hasConceptScore W2947706148C201995342 @default.
• W2947706148 hasConceptScore W2947706148C202887219 @default.
• W2947706148 hasConceptScore W2947706148C206588197 @default.
• W2947706148 hasConceptScore W2947706148C2776401178 @default.
• W2947706148 hasConceptScore W2947706148C2780451532 @default.
• W2947706148 hasConceptScore W2947706148C2780586882 @default.
• W2947706148 hasConceptScore W2947706148C2992317946 @default.

• next