Matches in SemOpenAlex for { <https://semopenalex.org/work/W3129120523> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W3129120523 abstract "Convolutional neural networks (CNNs) have achieved remarkable success in image recognition. Although the internal patterns of the input images are effectively learned by the CNNs, these patterns only constitute a small proportion of useful patterns contained in the input images. This can be attributed to the fact that the CNNs will stop learning if the learned patterns are enough to make a correct classification. Network regularization methods like dropout and SpatialDropout can ease this problem. During training, they randomly drop the features. These dropout methods, in essence, change the patterns learned by the networks, and in turn, forces the networks to learn other patterns to make the correct classification. However, the above methods have an important drawback. Randomly dropping features is generally inefficient and can introduce unnecessary noise. To tackle this problem, we propose SelectScale. Instead of randomly dropping units, SelectScale selects the important features in networks and adjusts them during training. Using SelectScale, we improve the performance of CNNs on CIFAR and ImageNet." @default.
- W3129120523 created "2021-02-15" @default.
- W3129120523 creator A5042114598 @default.
- W3129120523 creator A5050393292 @default.
- W3129120523 creator A5051862250 @default.
- W3129120523 creator A5053316728 @default.
- W3129120523 date "2020-11-30" @default.
- W3129120523 modified "2023-09-25" @default.
- W3129120523 title "SelectScale: Mining More Patterns from Images via Selective and Soft Dropout" @default.
- W3129120523 cites W1686810756 @default.
- W3129120523 cites W1903029394 @default.
- W3129120523 cites W2102605133 @default.
- W3129120523 cites W2117539524 @default.
- W3129120523 cites W2194775991 @default.
- W3129120523 cites W2295107390 @default.
- W3129120523 cites W2302255633 @default.
- W3129120523 cites W2401231614 @default.
- W3129120523 cites W2511730936 @default.
- W3129120523 cites W2518108298 @default.
- W3129120523 cites W2746314669 @default.
- W3129120523 cites W2781596748 @default.
- W3129120523 cites W2949117887 @default.
- W3129120523 cites W2949892913 @default.
- W3129120523 cites W2952634764 @default.
- W3129120523 cites W2953106684 @default.
- W3129120523 cites W2953328958 @default.
- W3129120523 doi "https://doi.org/10.48550/arxiv.2012.15766" @default.
- W3129120523 hasPublicationYear "2020" @default.
- W3129120523 type Work @default.
- W3129120523 sameAs 3129120523 @default.
- W3129120523 citedByCount "0" @default.
- W3129120523 crossrefType "posted-content" @default.
- W3129120523 hasAuthorship W3129120523A5042114598 @default.
- W3129120523 hasAuthorship W3129120523A5050393292 @default.
- W3129120523 hasAuthorship W3129120523A5051862250 @default.
- W3129120523 hasAuthorship W3129120523A5053316728 @default.
- W3129120523 hasBestOaLocation W31291205231 @default.
- W3129120523 hasConcept C115961682 @default.
- W3129120523 hasConcept C119857082 @default.
- W3129120523 hasConcept C153180895 @default.
- W3129120523 hasConcept C154945302 @default.
- W3129120523 hasConcept C2776135515 @default.
- W3129120523 hasConcept C2776145597 @default.
- W3129120523 hasConcept C41008148 @default.
- W3129120523 hasConcept C50644808 @default.
- W3129120523 hasConcept C75294576 @default.
- W3129120523 hasConcept C81363708 @default.
- W3129120523 hasConceptScore W3129120523C115961682 @default.
- W3129120523 hasConceptScore W3129120523C119857082 @default.
- W3129120523 hasConceptScore W3129120523C153180895 @default.
- W3129120523 hasConceptScore W3129120523C154945302 @default.
- W3129120523 hasConceptScore W3129120523C2776135515 @default.
- W3129120523 hasConceptScore W3129120523C2776145597 @default.
- W3129120523 hasConceptScore W3129120523C41008148 @default.
- W3129120523 hasConceptScore W3129120523C50644808 @default.
- W3129120523 hasConceptScore W3129120523C75294576 @default.
- W3129120523 hasConceptScore W3129120523C81363708 @default.
- W3129120523 hasLocation W31291205231 @default.
- W3129120523 hasOpenAccess W3129120523 @default.
- W3129120523 hasPrimaryLocation W31291205231 @default.
- W3129120523 hasRelatedWork W12336802 @default.
- W3129120523 hasRelatedWork W1383942 @default.
- W3129120523 hasRelatedWork W15119441 @default.
- W3129120523 hasRelatedWork W1562032 @default.
- W3129120523 hasRelatedWork W21705 @default.
- W3129120523 hasRelatedWork W4275953 @default.
- W3129120523 hasRelatedWork W4919037 @default.
- W3129120523 hasRelatedWork W6680660 @default.
- W3129120523 hasRelatedWork W9190101 @default.
- W3129120523 hasRelatedWork W9362070 @default.
- W3129120523 isParatext "false" @default.
- W3129120523 isRetracted "false" @default.
- W3129120523 magId "3129120523" @default.
- W3129120523 workType "article" @default.