Matches in SemOpenAlex for { <https://semopenalex.org/work/W2787697768> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W2787697768 endingPage "516" @default.
- W2787697768 startingPage "502" @default.
- W2787697768 abstract "In this paper, we introduce a method to compress intermediate feature maps of deep neural networks (DNNs) to decrease memory storage and bandwidth requirements during inference. Unlike previous works, the proposed method is based on converting fixed-point activations into vectors over the smallest GF(2) finite field followed by nonlinear dimensionality reduction (NDR) layers embedded into a DNN. Such an end-to-end learned representation finds more compact feature maps by exploiting quantization redundancies within the fixed-point activations along the channel or spatial dimensions. We apply the proposed network architectures derived from modified SqueezeNet and MobileNetV2 to the tasks of ImageNet classification and PASCAL VOC object detection. Compared to prior approaches, the conducted experiments show a factor of 2 decrease in memory requirements with minor degradation in accuracy while adding only bitwise computations." @default.
- W2787697768 created "2018-02-23" @default.
- W2787697768 creator A5026912892 @default.
- W2787697768 creator A5031821053 @default.
- W2787697768 creator A5062981565 @default.
- W2787697768 date "2019-01-01" @default.
- W2787697768 modified "2023-09-23" @default.
- W2787697768 title "DNN Feature Map Compression Using Learned Representation over GF(2)" @default.
- W2787697768 cites W1999085092 @default.
- W2787697768 cites W2031489346 @default.
- W2787697768 cites W2100495367 @default.
- W2787697768 cites W2117539524 @default.
- W2787697768 cites W2155893237 @default.
- W2787697768 cites W2194775991 @default.
- W2787697768 cites W2300242332 @default.
- W2787697768 cites W2557728737 @default.
- W2787697768 cites W2604700561 @default.
- W2787697768 cites W2963163009 @default.
- W2787697768 cites W2963182550 @default.
- W2787697768 cites W3106250896 @default.
- W2787697768 cites W4236868170 @default.
- W2787697768 cites W4249932213 @default.
- W2787697768 doi "https://doi.org/10.1007/978-3-030-11018-5_41" @default.
- W2787697768 hasPublicationYear "2019" @default.
- W2787697768 type Work @default.
- W2787697768 sameAs 2787697768 @default.
- W2787697768 citedByCount "8" @default.
- W2787697768 countsByYear W27876977682019 @default.
- W2787697768 countsByYear W27876977682021 @default.
- W2787697768 countsByYear W27876977682022 @default.
- W2787697768 crossrefType "book-chapter" @default.
- W2787697768 hasAuthorship W2787697768A5026912892 @default.
- W2787697768 hasAuthorship W2787697768A5031821053 @default.
- W2787697768 hasAuthorship W2787697768A5062981565 @default.
- W2787697768 hasBestOaLocation W27876977682 @default.
- W2787697768 hasConcept C138885662 @default.
- W2787697768 hasConcept C153180895 @default.
- W2787697768 hasConcept C154945302 @default.
- W2787697768 hasConcept C159985019 @default.
- W2787697768 hasConcept C17744445 @default.
- W2787697768 hasConcept C180016635 @default.
- W2787697768 hasConcept C192562407 @default.
- W2787697768 hasConcept C199539241 @default.
- W2787697768 hasConcept C2776359362 @default.
- W2787697768 hasConcept C2776401178 @default.
- W2787697768 hasConcept C41008148 @default.
- W2787697768 hasConcept C41895202 @default.
- W2787697768 hasConcept C94625758 @default.
- W2787697768 hasConceptScore W2787697768C138885662 @default.
- W2787697768 hasConceptScore W2787697768C153180895 @default.
- W2787697768 hasConceptScore W2787697768C154945302 @default.
- W2787697768 hasConceptScore W2787697768C159985019 @default.
- W2787697768 hasConceptScore W2787697768C17744445 @default.
- W2787697768 hasConceptScore W2787697768C180016635 @default.
- W2787697768 hasConceptScore W2787697768C192562407 @default.
- W2787697768 hasConceptScore W2787697768C199539241 @default.
- W2787697768 hasConceptScore W2787697768C2776359362 @default.
- W2787697768 hasConceptScore W2787697768C2776401178 @default.
- W2787697768 hasConceptScore W2787697768C41008148 @default.
- W2787697768 hasConceptScore W2787697768C41895202 @default.
- W2787697768 hasConceptScore W2787697768C94625758 @default.
- W2787697768 hasLocation W27876977681 @default.
- W2787697768 hasLocation W27876977682 @default.
- W2787697768 hasOpenAccess W2787697768 @default.
- W2787697768 hasPrimaryLocation W27876977681 @default.
- W2787697768 hasRelatedWork W2015538044 @default.
- W2787697768 hasRelatedWork W2016461833 @default.
- W2787697768 hasRelatedWork W2052253960 @default.
- W2787697768 hasRelatedWork W2147802381 @default.
- W2787697768 hasRelatedWork W2382607599 @default.
- W2787697768 hasRelatedWork W2489255581 @default.
- W2787697768 hasRelatedWork W2760085659 @default.
- W2787697768 hasRelatedWork W2787306535 @default.
- W2787697768 hasRelatedWork W3197541072 @default.
- W2787697768 hasRelatedWork W2480412556 @default.
- W2787697768 isParatext "false" @default.
- W2787697768 isRetracted "false" @default.
- W2787697768 magId "2787697768" @default.
- W2787697768 workType "book-chapter" @default.