Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385694585> ?p ?o ?g. }
Showing items 1 to 68 of
68
with 100 items per page.
- W4385694585 abstract "Image filters play a crucial role in the performance of convolutional neural networks (CNNs). Yet, the optimisation of those filters tends to focus solely on optimising their weights. As a result, the machine learning practitioner either uses the standard 3×3 filter size or has to select the filter size empirically, as the optimal filter size depends on the application at hand. There has been a lack of serious attempts to address this issue in CNNs. To this end, we propose a novel technique to learn the filter size without depending on either gradient descent or backpropagation. We compare our technique with the only serious attempt existing and show that not only our method performs better, but also converges to an optimal solution at a much smaller number of training iterations." @default.
- W4385694585 created "2023-08-10" @default.
- W4385694585 creator A5004587843 @default.
- W4385694585 creator A5045981758 @default.
- W4385694585 creator A5052748715 @default.
- W4385694585 date "2023-07-02" @default.
- W4385694585 modified "2023-09-27" @default.
- W4385694585 title "A novel technique for optimizing the filter size of CNNs without backpropagation" @default.
- W4385694585 cites W2058262254 @default.
- W4385694585 cites W2097117768 @default.
- W4385694585 cites W2117539524 @default.
- W4385694585 cites W2194775991 @default.
- W4385694585 cites W2243226955 @default.
- W4385694585 cites W2961719374 @default.
- W4385694585 cites W2964140963 @default.
- W4385694585 cites W3198949520 @default.
- W4385694585 doi "https://doi.org/10.1109/ssp53291.2023.10207944" @default.
- W4385694585 hasPublicationYear "2023" @default.
- W4385694585 type Work @default.
- W4385694585 citedByCount "0" @default.
- W4385694585 crossrefType "proceedings-article" @default.
- W4385694585 hasAuthorship W4385694585A5004587843 @default.
- W4385694585 hasAuthorship W4385694585A5045981758 @default.
- W4385694585 hasAuthorship W4385694585A5052748715 @default.
- W4385694585 hasConcept C106131492 @default.
- W4385694585 hasConcept C11413529 @default.
- W4385694585 hasConcept C119857082 @default.
- W4385694585 hasConcept C120665830 @default.
- W4385694585 hasConcept C121332964 @default.
- W4385694585 hasConcept C153180895 @default.
- W4385694585 hasConcept C153258448 @default.
- W4385694585 hasConcept C154945302 @default.
- W4385694585 hasConcept C155032097 @default.
- W4385694585 hasConcept C192209626 @default.
- W4385694585 hasConcept C31972630 @default.
- W4385694585 hasConcept C41008148 @default.
- W4385694585 hasConcept C50644808 @default.
- W4385694585 hasConcept C81363708 @default.
- W4385694585 hasConceptScore W4385694585C106131492 @default.
- W4385694585 hasConceptScore W4385694585C11413529 @default.
- W4385694585 hasConceptScore W4385694585C119857082 @default.
- W4385694585 hasConceptScore W4385694585C120665830 @default.
- W4385694585 hasConceptScore W4385694585C121332964 @default.
- W4385694585 hasConceptScore W4385694585C153180895 @default.
- W4385694585 hasConceptScore W4385694585C153258448 @default.
- W4385694585 hasConceptScore W4385694585C154945302 @default.
- W4385694585 hasConceptScore W4385694585C155032097 @default.
- W4385694585 hasConceptScore W4385694585C192209626 @default.
- W4385694585 hasConceptScore W4385694585C31972630 @default.
- W4385694585 hasConceptScore W4385694585C41008148 @default.
- W4385694585 hasConceptScore W4385694585C50644808 @default.
- W4385694585 hasConceptScore W4385694585C81363708 @default.
- W4385694585 hasLocation W43856945851 @default.
- W4385694585 hasOpenAccess W4385694585 @default.
- W4385694585 hasPrimaryLocation W43856945851 @default.
- W4385694585 hasRelatedWork W1183256782 @default.
- W4385694585 hasRelatedWork W1979695132 @default.
- W4385694585 hasRelatedWork W2391384657 @default.
- W4385694585 hasRelatedWork W2767651786 @default.
- W4385694585 hasRelatedWork W2912288872 @default.
- W4385694585 hasRelatedWork W3021430260 @default.
- W4385694585 hasRelatedWork W3027997911 @default.
- W4385694585 hasRelatedWork W3123071383 @default.
- W4385694585 hasRelatedWork W3159389381 @default.
- W4385694585 hasRelatedWork W4287776258 @default.
- W4385694585 isParatext "false" @default.
- W4385694585 isRetracted "false" @default.
- W4385694585 workType "article" @default.