FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951283089> ?p ?o ?g. }

• W2951283089 abstract "Symmetry is present in nature and science. In image processing, kernels for spatial filtering possess some symmetry (e.g. Sobel operators, Gaussian, Laplacian). Convolutional layers in artificial feed-forward neural networks have typically considered the kernel weights without any constraint. In this paper, we propose to investigate the impact of a symmetry constraint in convolutional layers for image classification tasks, taking our inspiration from the processes involved in the primary visual cortex and common image processing techniques. The goal is to assess the extent to which it is possible to enforce symmetrical constraints on the filters throughout the training process of a convolutional neural network (CNN) by modifying the weight update preformed during the backpropagation algorithm and to evaluate the change in performance. The main hypothesis of this paper is that the symmetrical constraint reduces the number of free parameters in the network, and it is able to achieve near identical performance to the modern methodology of training. In particular, we address the following cases: x/y-axis symmetry, point reflection, and anti-point reflection. The performance has been evaluated on four databases of images. The results support the conclusion that while random weights offer more freedom to the model, the symmetry constraint provides a similar level of performance while decreasing substantially the number of free parameters in the model. Such an approach can be valuable in phase-sensitive applications that require a linear phase property throughout the feature extraction process." @default.
• W2951283089 created "2019-06-27" @default.
• W2951283089 creator A5010475134 @default.
• W2951283089 creator A5025447862 @default.
• W2951283089 date "2019-06-10" @default.
• W2951283089 modified "2023-09-27" @default.
• W2951283089 title "SymNet: Symmetrical Filters in Convolutional Neural Networks." @default.
• W2951283089 cites W1499677203 @default.
• W2951283089 cites W1766888123 @default.
• W2951283089 cites W1892339738 @default.
• W2951283089 cites W1969840923 @default.
• W2951283089 cites W1986216697 @default.
• W2951283089 cites W1999798000 @default.
• W2951283089 cites W2001695906 @default.
• W2951283089 cites W2022485710 @default.
• W2951283089 cites W2031878488 @default.
• W2951283089 cites W2044738244 @default.
• W2951283089 cites W2044943596 @default.
• W2951283089 cites W2061065224 @default.
• W2951283089 cites W2095705004 @default.
• W2951283089 cites W2097117768 @default.
• W2951283089 cites W2101926813 @default.
• W2951283089 cites W2111072639 @default.
• W2951283089 cites W2112162870 @default.
• W2951283089 cites W2112796928 @default.
• W2951283089 cites W2116216752 @default.
• W2951283089 cites W2122410182 @default.
• W2951283089 cites W2124386111 @default.
• W2951283089 cites W2132032936 @default.
• W2951283089 cites W2133059825 @default.
• W2951283089 cites W2138857742 @default.
• W2951283089 cites W2140776489 @default.
• W2951283089 cites W2141125852 @default.
• W2951283089 cites W2149194912 @default.
• W2951283089 cites W2150590430 @default.
• W2951283089 cites W2151335099 @default.
• W2951283089 cites W2153156486 @default.
• W2951283089 cites W2154579312 @default.
• W2951283089 cites W2156163116 @default.
• W2951283089 cites W2156700398 @default.
• W2951283089 cites W2163605009 @default.
• W2951283089 cites W2499468060 @default.
• W2951283089 cites W2545058703 @default.
• W2951283089 cites W2546302380 @default.
• W2951283089 cites W2807235401 @default.
• W2951283089 cites W2962835968 @default.
• W2951283089 cites W3147875956 @default.
• W2951283089 cites W3150925829 @default.
• W2951283089 cites W571345558 @default.
• W2951283089 hasPublicationYear "2019" @default.
• W2951283089 type Work @default.
• W2951283089 sameAs 2951283089 @default.
• W2951283089 citedByCount "0" @default.
• W2951283089 crossrefType "posted-content" @default.
• W2951283089 hasAuthorship W2951283089A5010475134 @default.
• W2951283089 hasAuthorship W2951283089A5025447862 @default.
• W2951283089 hasConcept C11413529 @default.
• W2951283089 hasConcept C118615104 @default.
• W2951283089 hasConcept C138885662 @default.
• W2951283089 hasConcept C153180895 @default.
• W2951283089 hasConcept C154945302 @default.
• W2951283089 hasConcept C155032097 @default.
• W2951283089 hasConcept C2524010 @default.
• W2951283089 hasConcept C2776036281 @default.
• W2951283089 hasConcept C2776401178 @default.
• W2951283089 hasConcept C2779886137 @default.
• W2951283089 hasConcept C33923547 @default.
• W2951283089 hasConcept C41008148 @default.
• W2951283089 hasConcept C41895202 @default.
• W2951283089 hasConcept C50644808 @default.
• W2951283089 hasConcept C52622490 @default.
• W2951283089 hasConcept C74193536 @default.
• W2951283089 hasConcept C81363708 @default.
• W2951283089 hasConceptScore W2951283089C11413529 @default.
• W2951283089 hasConceptScore W2951283089C118615104 @default.
• W2951283089 hasConceptScore W2951283089C138885662 @default.
• W2951283089 hasConceptScore W2951283089C153180895 @default.
• W2951283089 hasConceptScore W2951283089C154945302 @default.
• W2951283089 hasConceptScore W2951283089C155032097 @default.
• W2951283089 hasConceptScore W2951283089C2524010 @default.
• W2951283089 hasConceptScore W2951283089C2776036281 @default.
• W2951283089 hasConceptScore W2951283089C2776401178 @default.
• W2951283089 hasConceptScore W2951283089C2779886137 @default.
• W2951283089 hasConceptScore W2951283089C33923547 @default.
• W2951283089 hasConceptScore W2951283089C41008148 @default.
• W2951283089 hasConceptScore W2951283089C41895202 @default.
• W2951283089 hasConceptScore W2951283089C50644808 @default.
• W2951283089 hasConceptScore W2951283089C52622490 @default.
• W2951283089 hasConceptScore W2951283089C74193536 @default.
• W2951283089 hasConceptScore W2951283089C81363708 @default.
• W2951283089 hasLocation W29512830891 @default.
• W2951283089 hasOpenAccess W2951283089 @default.
• W2951283089 hasPrimaryLocation W29512830891 @default.
• W2951283089 hasRelatedWork W2106224199 @default.
• W2951283089 hasRelatedWork W2413005285 @default.
• W2951283089 hasRelatedWork W2554242204 @default.
• W2951283089 hasRelatedWork W2593110912 @default.
• W2951283089 hasRelatedWork W2785631679 @default.
• W2951283089 hasRelatedWork W2805022637 @default.
• W2951283089 hasRelatedWork W2806226713 @default.

• next