Matches in SemOpenAlex for { <https://semopenalex.org/work/W3021330351> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3021330351 abstract "Traditionally, deep convolutional neural networks consist of a series of convolutional and pooling layers followed by one or more fully connected (FC) layers to perform the final classification. While this design has been successful, for datasets with a large number of categories, the fully connected layers often account for a large percentage of the network's parameters. For applications with memory constraints, such as mobile devices and embedded platforms, this is not ideal. Recently, a family of architectures that involve replacing the learned fully connected output layer with a fixed layer has been proposed as a way to achieve better efficiency. In this paper we examine this idea further and demonstrate that fixed classifiers offer no additional benefit compared to simply removing the output layer along with its parameters. We further demonstrate that the typical approach of having a fully connected final output layer is inefficient in terms of parameter count. We are able to achieve comparable performance to a traditionally learned fully connected classification output layer on the ImageNet-1K, CIFAR-100, Stanford Cars-196, and Oxford Flowers-102 datasets, while not having a fully connected output layer at all." @default.
- W3021330351 created "2020-05-13" @default.
- W3021330351 creator A5030503283 @default.
- W3021330351 creator A5035388148 @default.
- W3021330351 creator A5046979072 @default.
- W3021330351 creator A5047193437 @default.
- W3021330351 date "2020-04-28" @default.
- W3021330351 modified "2023-09-27" @default.
- W3021330351 title "Do We Need Fully Connected Output Layers in Convolutional Networks?" @default.
- W3021330351 cites W2114766824 @default.
- W3021330351 cites W2117539524 @default.
- W3021330351 cites W2123045220 @default.
- W3021330351 cites W2138011018 @default.
- W3021330351 cites W2163605009 @default.
- W3021330351 cites W2194775991 @default.
- W3021330351 cites W2266701264 @default.
- W3021330351 cites W2279098554 @default.
- W3021330351 cites W2295107390 @default.
- W3021330351 cites W2319920447 @default.
- W3021330351 cites W2495425901 @default.
- W3021330351 cites W2531409750 @default.
- W3021330351 cites W2533598788 @default.
- W3021330351 cites W2554931888 @default.
- W3021330351 cites W2593110912 @default.
- W3021330351 cites W2771964490 @default.
- W3021330351 cites W2788388592 @default.
- W3021330351 cites W2883780447 @default.
- W3021330351 cites W2920160940 @default.
- W3021330351 cites W2962965870 @default.
- W3021330351 cites W2963103976 @default.
- W3021330351 cites W2963122961 @default.
- W3021330351 cites W2963163009 @default.
- W3021330351 cites W2963446712 @default.
- W3021330351 cites W2963674932 @default.
- W3021330351 cites W2963828549 @default.
- W3021330351 cites W2963940260 @default.
- W3021330351 cites W2967601231 @default.
- W3021330351 cites W2968596670 @default.
- W3021330351 cites W2971524484 @default.
- W3021330351 cites W3118608800 @default.
- W3021330351 cites W992687842 @default.
- W3021330351 doi "https://doi.org/10.48550/arxiv.2004.13587" @default.
- W3021330351 hasPublicationYear "2020" @default.
- W3021330351 type Work @default.
- W3021330351 sameAs 3021330351 @default.
- W3021330351 citedByCount "1" @default.
- W3021330351 countsByYear W30213303512020 @default.
- W3021330351 crossrefType "posted-content" @default.
- W3021330351 hasAuthorship W3021330351A5030503283 @default.
- W3021330351 hasAuthorship W3021330351A5035388148 @default.
- W3021330351 hasAuthorship W3021330351A5046979072 @default.
- W3021330351 hasAuthorship W3021330351A5047193437 @default.
- W3021330351 hasBestOaLocation W30213303511 @default.
- W3021330351 hasConcept C111472728 @default.
- W3021330351 hasConcept C138885662 @default.
- W3021330351 hasConcept C153180895 @default.
- W3021330351 hasConcept C154945302 @default.
- W3021330351 hasConcept C178790620 @default.
- W3021330351 hasConcept C185592680 @default.
- W3021330351 hasConcept C2776639384 @default.
- W3021330351 hasConcept C2779227376 @default.
- W3021330351 hasConcept C41008148 @default.
- W3021330351 hasConcept C70437156 @default.
- W3021330351 hasConcept C81363708 @default.
- W3021330351 hasConceptScore W3021330351C111472728 @default.
- W3021330351 hasConceptScore W3021330351C138885662 @default.
- W3021330351 hasConceptScore W3021330351C153180895 @default.
- W3021330351 hasConceptScore W3021330351C154945302 @default.
- W3021330351 hasConceptScore W3021330351C178790620 @default.
- W3021330351 hasConceptScore W3021330351C185592680 @default.
- W3021330351 hasConceptScore W3021330351C2776639384 @default.
- W3021330351 hasConceptScore W3021330351C2779227376 @default.
- W3021330351 hasConceptScore W3021330351C41008148 @default.
- W3021330351 hasConceptScore W3021330351C70437156 @default.
- W3021330351 hasConceptScore W3021330351C81363708 @default.
- W3021330351 hasLocation W30213303511 @default.
- W3021330351 hasOpenAccess W3021330351 @default.
- W3021330351 hasPrimaryLocation W30213303511 @default.
- W3021330351 hasRelatedWork W2368694199 @default.
- W3021330351 hasRelatedWork W2424871898 @default.
- W3021330351 hasRelatedWork W2514274290 @default.
- W3021330351 hasRelatedWork W2517027266 @default.
- W3021330351 hasRelatedWork W2613736958 @default.
- W3021330351 hasRelatedWork W2758063741 @default.
- W3021330351 hasRelatedWork W2792080776 @default.
- W3021330351 hasRelatedWork W2951391129 @default.
- W3021330351 hasRelatedWork W2969680539 @default.
- W3021330351 hasRelatedWork W3093612317 @default.
- W3021330351 isParatext "false" @default.
- W3021330351 isRetracted "false" @default.
- W3021330351 magId "3021330351" @default.
- W3021330351 workType "article" @default.