Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949273893> ?p ?o ?g. }
- W2949273893 abstract "We revisit the idea of brain damage, i.e. the pruning of the coefficients of a neural network, and suggest how brain damage can be modified and used to speedup convolutional layers. The approach uses the fact that many efficient implementations reduce generalized convolutions to matrix multiplications. The suggested brain damage process prunes the convolutional kernel tensor in a group-wise fashion by adding group-sparsity regularization to the standard training process. After such group-wise pruning, convolutions can be reduced to multiplications of thinned dense matrices, which leads to speedup. In the comparison on AlexNet, the method achieves very competitive performance." @default.
- W2949273893 created "2019-06-27" @default.
- W2949273893 creator A5025519425 @default.
- W2949273893 creator A5083458209 @default.
- W2949273893 date "2015-06-08" @default.
- W2949273893 modified "2023-09-27" @default.
- W2949273893 title "Fast ConvNets Using Group-wise Brain Damage" @default.
- W2949273893 cites W1530262073 @default.
- W2949273893 cites W1556531089 @default.
- W2949273893 cites W1570197553 @default.
- W2949273893 cites W1667652561 @default.
- W2949273893 cites W1821462560 @default.
- W2949273893 cites W1841592590 @default.
- W2949273893 cites W1920702274 @default.
- W2949273893 cites W1922123711 @default.
- W2949273893 cites W1932847118 @default.
- W2949273893 cites W1984915212 @default.
- W2949273893 cites W2002648693 @default.
- W2949273893 cites W2108687351 @default.
- W2949273893 cites W2112796928 @default.
- W2949273893 cites W2114766824 @default.
- W2949273893 cites W2133257461 @default.
- W2949273893 cites W2134797427 @default.
- W2949273893 cites W2135017741 @default.
- W2949273893 cites W2138019504 @default.
- W2949273893 cites W2141362318 @default.
- W2949273893 cites W2147800946 @default.
- W2949273893 cites W2149381887 @default.
- W2949273893 cites W2155541015 @default.
- W2949273893 cites W2167215970 @default.
- W2949273893 cites W2204975001 @default.
- W2949273893 cites W2290703437 @default.
- W2949273893 cites W2294543795 @default.
- W2949273893 cites W2949234772 @default.
- W2949273893 cites W2950094539 @default.
- W2949273893 cites W2950967261 @default.
- W2949273893 cites W2952432176 @default.
- W2949273893 cites W2952689122 @default.
- W2949273893 cites W2953066166 @default.
- W2949273893 cites W2962835968 @default.
- W2949273893 cites W2963367891 @default.
- W2949273893 cites W2964118293 @default.
- W2949273893 hasPublicationYear "2015" @default.
- W2949273893 type Work @default.
- W2949273893 sameAs 2949273893 @default.
- W2949273893 citedByCount "19" @default.
- W2949273893 countsByYear W29492738932015 @default.
- W2949273893 countsByYear W29492738932016 @default.
- W2949273893 countsByYear W29492738932017 @default.
- W2949273893 countsByYear W29492738932018 @default.
- W2949273893 countsByYear W29492738932019 @default.
- W2949273893 countsByYear W29492738932020 @default.
- W2949273893 countsByYear W29492738932021 @default.
- W2949273893 crossrefType "posted-content" @default.
- W2949273893 hasAuthorship W2949273893A5025519425 @default.
- W2949273893 hasAuthorship W2949273893A5083458209 @default.
- W2949273893 hasConcept C108010975 @default.
- W2949273893 hasConcept C11413529 @default.
- W2949273893 hasConcept C114614502 @default.
- W2949273893 hasConcept C121332964 @default.
- W2949273893 hasConcept C153180895 @default.
- W2949273893 hasConcept C154945302 @default.
- W2949273893 hasConcept C155281189 @default.
- W2949273893 hasConcept C173608175 @default.
- W2949273893 hasConcept C202444582 @default.
- W2949273893 hasConcept C2776135515 @default.
- W2949273893 hasConcept C2781311116 @default.
- W2949273893 hasConcept C33923547 @default.
- W2949273893 hasConcept C41008148 @default.
- W2949273893 hasConcept C62520636 @default.
- W2949273893 hasConcept C6557445 @default.
- W2949273893 hasConcept C68339613 @default.
- W2949273893 hasConcept C74193536 @default.
- W2949273893 hasConcept C81363708 @default.
- W2949273893 hasConcept C86803240 @default.
- W2949273893 hasConceptScore W2949273893C108010975 @default.
- W2949273893 hasConceptScore W2949273893C11413529 @default.
- W2949273893 hasConceptScore W2949273893C114614502 @default.
- W2949273893 hasConceptScore W2949273893C121332964 @default.
- W2949273893 hasConceptScore W2949273893C153180895 @default.
- W2949273893 hasConceptScore W2949273893C154945302 @default.
- W2949273893 hasConceptScore W2949273893C155281189 @default.
- W2949273893 hasConceptScore W2949273893C173608175 @default.
- W2949273893 hasConceptScore W2949273893C202444582 @default.
- W2949273893 hasConceptScore W2949273893C2776135515 @default.
- W2949273893 hasConceptScore W2949273893C2781311116 @default.
- W2949273893 hasConceptScore W2949273893C33923547 @default.
- W2949273893 hasConceptScore W2949273893C41008148 @default.
- W2949273893 hasConceptScore W2949273893C62520636 @default.
- W2949273893 hasConceptScore W2949273893C6557445 @default.
- W2949273893 hasConceptScore W2949273893C68339613 @default.
- W2949273893 hasConceptScore W2949273893C74193536 @default.
- W2949273893 hasConceptScore W2949273893C81363708 @default.
- W2949273893 hasConceptScore W2949273893C86803240 @default.
- W2949273893 hasLocation W29492738931 @default.
- W2949273893 hasOpenAccess W2949273893 @default.
- W2949273893 hasPrimaryLocation W29492738931 @default.
- W2949273893 hasRelatedWork W1570197553 @default.
- W2949273893 hasRelatedWork W1686810756 @default.
- W2949273893 hasRelatedWork W1724438581 @default.