FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3179567590> ?p ?o ?g. }

• W3179567590 abstract "The square kernel is a standard unit for contemporary Convolutional Neural Networks (CNNs), as it fits well on the tensor computation for the convolution operation. However, the receptive field in the human visual system is actually isotropic like a circle. Motivated by this observation, we propose using circle kernels with isotropic receptive fields for the convolution, and our training takes approximately equivalent amount of calculation when compared with the corresponding CNN with square kernels. Our preliminary experiments demonstrate the rationality of circle kernels. We then propose a kernel boosting strategy that integrates the circle kernels with square kernels for the training and inference, and we further let the kernel size/radius be learnable during the training. Note that we reparameterize the circle kernels or integrated kernels before the inference, thus taking no extra computation as well as the number of parameter overhead for the testing. Extensive experiments on several standard datasets, ImageNet, CIFAR-10 and CIFAR-100, using the circle kernels or integrated kernels on typical existing CNNs, show that our approach exhibits highly competitive performance. Specifically, on ImageNet with standard data augmentation, our approach dramatically boosts the performance of MobileNetV3-Small by 5.20% top-1 accuracy and 3.39% top-5 accuracy, and boosts the performance of MobileNetV3-Large by 2.16% top-1 accuracy and 1.18% top-5 accuracy." @default.
• W3179567590 created "2021-07-19" @default.
• W3179567590 creator A5013240918 @default.
• W3179567590 creator A5041130370 @default.
• W3179567590 creator A5043211393 @default.
• W3179567590 creator A5046831575 @default.
• W3179567590 creator A5079036364 @default.
• W3179567590 date "2021-07-06" @default.
• W3179567590 modified "2023-09-27" @default.
• W3179567590 title "Integrating Circle Kernels into Convolutional Neural Networks." @default.
• W3179567590 cites W1593114658 @default.
• W3179567590 cites W1686810756 @default.
• W3179567590 cites W2108598243 @default.
• W3179567590 cites W2112796928 @default.
• W3179567590 cites W2150884987 @default.
• W3179567590 cites W2194775991 @default.
• W3179567590 cites W2302255633 @default.
• W3179567590 cites W2331143823 @default.
• W3179567590 cites W2401231614 @default.
• W3179567590 cites W2556967412 @default.
• W3179567590 cites W2601564443 @default.
• W3179567590 cites W2612445135 @default.
• W3179567590 cites W2626639386 @default.
• W3179567590 cites W2746314669 @default.
• W3179567590 cites W2932077855 @default.
• W3179567590 cites W2953604046 @default.
• W3179567590 cites W2963446712 @default.
• W3179567590 cites W2963674932 @default.
• W3179567590 cites W2963821229 @default.
• W3179567590 cites W2963975324 @default.
• W3179567590 cites W2964081807 @default.
• W3179567590 cites W2964259004 @default.
• W3179567590 cites W2966926453 @default.
• W3179567590 cites W2995645057 @default.
• W3179567590 cites W3109403878 @default.
• W3179567590 cites W3118608800 @default.
• W3179567590 cites W3120735991 @default.
• W3179567590 cites W3164008977 @default.
• W3179567590 hasPublicationYear "2021" @default.
• W3179567590 type Work @default.
• W3179567590 sameAs 3179567590 @default.
• W3179567590 citedByCount "0" @default.
• W3179567590 crossrefType "posted-content" @default.
• W3179567590 hasAuthorship W3179567590A5013240918 @default.
• W3179567590 hasAuthorship W3179567590A5041130370 @default.
• W3179567590 hasAuthorship W3179567590A5043211393 @default.
• W3179567590 hasAuthorship W3179567590A5046831575 @default.
• W3179567590 hasAuthorship W3179567590A5079036364 @default.
• W3179567590 hasConcept C11413529 @default.
• W3179567590 hasConcept C118615104 @default.
• W3179567590 hasConcept C126875415 @default.
• W3179567590 hasConcept C135692309 @default.
• W3179567590 hasConcept C153180895 @default.
• W3179567590 hasConcept C154945302 @default.
• W3179567590 hasConcept C2524010 @default.
• W3179567590 hasConcept C2776214188 @default.
• W3179567590 hasConcept C33923547 @default.
• W3179567590 hasConcept C41008148 @default.
• W3179567590 hasConcept C45347329 @default.
• W3179567590 hasConcept C45374587 @default.
• W3179567590 hasConcept C50644808 @default.
• W3179567590 hasConcept C74193536 @default.
• W3179567590 hasConcept C81363708 @default.
• W3179567590 hasConceptScore W3179567590C11413529 @default.
• W3179567590 hasConceptScore W3179567590C118615104 @default.
• W3179567590 hasConceptScore W3179567590C126875415 @default.
• W3179567590 hasConceptScore W3179567590C135692309 @default.
• W3179567590 hasConceptScore W3179567590C153180895 @default.
• W3179567590 hasConceptScore W3179567590C154945302 @default.
• W3179567590 hasConceptScore W3179567590C2524010 @default.
• W3179567590 hasConceptScore W3179567590C2776214188 @default.
• W3179567590 hasConceptScore W3179567590C33923547 @default.
• W3179567590 hasConceptScore W3179567590C41008148 @default.
• W3179567590 hasConceptScore W3179567590C45347329 @default.
• W3179567590 hasConceptScore W3179567590C45374587 @default.
• W3179567590 hasConceptScore W3179567590C50644808 @default.
• W3179567590 hasConceptScore W3179567590C74193536 @default.
• W3179567590 hasConceptScore W3179567590C81363708 @default.
• W3179567590 hasLocation W31795675901 @default.
• W3179567590 hasOpenAccess W3179567590 @default.
• W3179567590 hasPrimaryLocation W31795675901 @default.
• W3179567590 hasRelatedWork W2129511572 @default.
• W3179567590 hasRelatedWork W2165417234 @default.
• W3179567590 hasRelatedWork W2167085024 @default.
• W3179567590 hasRelatedWork W2202303789 @default.
• W3179567590 hasRelatedWork W2395739145 @default.
• W3179567590 hasRelatedWork W2514923279 @default.
• W3179567590 hasRelatedWork W2551469986 @default.
• W3179567590 hasRelatedWork W2591637645 @default.
• W3179567590 hasRelatedWork W2754377004 @default.
• W3179567590 hasRelatedWork W2768114338 @default.
• W3179567590 hasRelatedWork W2891297252 @default.
• W3179567590 hasRelatedWork W2914085889 @default.
• W3179567590 hasRelatedWork W2997905006 @default.
• W3179567590 hasRelatedWork W3013114425 @default.
• W3179567590 hasRelatedWork W3035433747 @default.
• W3179567590 hasRelatedWork W3095523882 @default.
• W3179567590 hasRelatedWork W3097466625 @default.
• W3179567590 hasRelatedWork W3100568443 @default.
• W3179567590 hasRelatedWork W3101069636 @default.

• next