Matches in SemOpenAlex for { <https://semopenalex.org/work/W4377710209> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4377710209 endingPage "15" @default.
- W4377710209 startingPage "1" @default.
- W4377710209 abstract "Observing that the existing model compression approaches only focus on reducing the redundancies in convolutional neural networks (CNNs) along one particular dimension (e.g., the channel or spatial or temporal dimension), in this work, we propose our multidimensional pruning (MDP) framework, which can compress both 2-D CNNs and 3-D CNNs along multiple dimensions in an end-to-end fashion. Specifically, MDP indicates the simultaneous reduction of channels and more redundancy on other additional dimensions. The redundancy of additional dimensions depends on the input data, i.e., spatial dimension for 2-D CNNs when using images as the input data, and spatial and temporal dimensions for 3-D CNNs when using videos as the input data. We further extend our MDP framework to the MDP-Point approach for compressing point cloud neural networks (PCNNs) whose inputs are irregular point clouds (e.g., PointNet). In this case, the redundancy along the additional dimension indicates the point dimension (i.e., the number of points). Comprehensive experiments on six benchmark datasets demonstrate the effectiveness of our MDP framework and its extended version MDP-Point for compressing CNNs and PCNNs, respectively." @default.
- W4377710209 created "2023-05-24" @default.
- W4377710209 creator A5039703932 @default.
- W4377710209 creator A5082181536 @default.
- W4377710209 creator A5087818121 @default.
- W4377710209 date "2023-01-01" @default.
- W4377710209 modified "2023-10-01" @default.
- W4377710209 title "Multidimensional Pruning and Its Extension: A Unified Framework for Model Compression" @default.
- W4377710209 doi "https://doi.org/10.1109/tnnls.2023.3266435" @default.
- W4377710209 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37220047" @default.
- W4377710209 hasPublicationYear "2023" @default.
- W4377710209 type Work @default.
- W4377710209 citedByCount "0" @default.
- W4377710209 crossrefType "journal-article" @default.
- W4377710209 hasAuthorship W4377710209A5039703932 @default.
- W4377710209 hasAuthorship W4377710209A5082181536 @default.
- W4377710209 hasAuthorship W4377710209A5087818121 @default.
- W4377710209 hasConcept C108010975 @default.
- W4377710209 hasConcept C111919701 @default.
- W4377710209 hasConcept C11413529 @default.
- W4377710209 hasConcept C131979681 @default.
- W4377710209 hasConcept C13280743 @default.
- W4377710209 hasConcept C152124472 @default.
- W4377710209 hasConcept C153180895 @default.
- W4377710209 hasConcept C154945302 @default.
- W4377710209 hasConcept C185798385 @default.
- W4377710209 hasConcept C202444582 @default.
- W4377710209 hasConcept C205649164 @default.
- W4377710209 hasConcept C33676613 @default.
- W4377710209 hasConcept C33923547 @default.
- W4377710209 hasConcept C41008148 @default.
- W4377710209 hasConcept C6557445 @default.
- W4377710209 hasConcept C70518039 @default.
- W4377710209 hasConcept C81363708 @default.
- W4377710209 hasConcept C86803240 @default.
- W4377710209 hasConceptScore W4377710209C108010975 @default.
- W4377710209 hasConceptScore W4377710209C111919701 @default.
- W4377710209 hasConceptScore W4377710209C11413529 @default.
- W4377710209 hasConceptScore W4377710209C131979681 @default.
- W4377710209 hasConceptScore W4377710209C13280743 @default.
- W4377710209 hasConceptScore W4377710209C152124472 @default.
- W4377710209 hasConceptScore W4377710209C153180895 @default.
- W4377710209 hasConceptScore W4377710209C154945302 @default.
- W4377710209 hasConceptScore W4377710209C185798385 @default.
- W4377710209 hasConceptScore W4377710209C202444582 @default.
- W4377710209 hasConceptScore W4377710209C205649164 @default.
- W4377710209 hasConceptScore W4377710209C33676613 @default.
- W4377710209 hasConceptScore W4377710209C33923547 @default.
- W4377710209 hasConceptScore W4377710209C41008148 @default.
- W4377710209 hasConceptScore W4377710209C6557445 @default.
- W4377710209 hasConceptScore W4377710209C70518039 @default.
- W4377710209 hasConceptScore W4377710209C81363708 @default.
- W4377710209 hasConceptScore W4377710209C86803240 @default.
- W4377710209 hasFunder F4320321001 @default.
- W4377710209 hasFunder F4320335777 @default.
- W4377710209 hasLocation W43777102091 @default.
- W4377710209 hasLocation W43777102092 @default.
- W4377710209 hasOpenAccess W4377710209 @default.
- W4377710209 hasPrimaryLocation W43777102091 @default.
- W4377710209 hasRelatedWork W2175746458 @default.
- W4377710209 hasRelatedWork W2732542196 @default.
- W4377710209 hasRelatedWork W2760085659 @default.
- W4377710209 hasRelatedWork W2767651786 @default.
- W4377710209 hasRelatedWork W2883200793 @default.
- W4377710209 hasRelatedWork W2949389737 @default.
- W4377710209 hasRelatedWork W2963556241 @default.
- W4377710209 hasRelatedWork W3018421652 @default.
- W4377710209 hasRelatedWork W3093612317 @default.
- W4377710209 hasRelatedWork W3094624326 @default.
- W4377710209 isParatext "false" @default.
- W4377710209 isRetracted "false" @default.
- W4377710209 workType "article" @default.