Matches in SemOpenAlex for { <https://semopenalex.org/work/W2988768032> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2988768032 abstract "Sequential data such as video are characterized by spatio-temporal correlations. As of yet, few deep learning algorithms exploit them to decrease the often massive cost during inference. This work leverages correlations in video data to linearize part of a deep neural network and thus reduce its size and computational cost. Drawing upon the simplicity of the typically used rectifier activation function, we replace the ReLU function by dynamically updating masks. The resulting layer stack is a simple chain of matrix multiplications and bias additions, that can be contracted into a single weight matrix and bias vector. Inference then reduces to an affine transformation of the input sequence with these contracted parameters. We show that the method is akin to approximating the neural network with a first-order Taylor expansion around a dynamically updating reference point. The proposed algorithm is evaluated on a denoising convolutional autoencoder." @default.
- W2988768032 created "2019-11-22" @default.
- W2988768032 creator A5001910622 @default.
- W2988768032 creator A5053821067 @default.
- W2988768032 date "2019-09-01" @default.
- W2988768032 modified "2023-10-03" @default.
- W2988768032 title "Linear Approximation of Deep Neural Networks for Efficient Inference on Video Data" @default.
- W2988768032 cites W2112796928 @default.
- W2988768032 cites W2195388612 @default.
- W2988768032 cites W2279098554 @default.
- W2988768032 cites W2291160084 @default.
- W2988768032 cites W2524428287 @default.
- W2988768032 cites W2552900565 @default.
- W2988768032 cites W2582144945 @default.
- W2988768032 cites W2591954064 @default.
- W2988768032 cites W2606822607 @default.
- W2988768032 cites W2612445135 @default.
- W2988768032 cites W2751477244 @default.
- W2988768032 cites W2962851944 @default.
- W2988768032 cites W2963727651 @default.
- W2988768032 cites W2964182247 @default.
- W2988768032 cites W2964299589 @default.
- W2988768032 doi "https://doi.org/10.23919/eusipco.2019.8902997" @default.
- W2988768032 hasPublicationYear "2019" @default.
- W2988768032 type Work @default.
- W2988768032 sameAs 2988768032 @default.
- W2988768032 citedByCount "2" @default.
- W2988768032 countsByYear W29887680322020 @default.
- W2988768032 countsByYear W29887680322022 @default.
- W2988768032 crossrefType "proceedings-article" @default.
- W2988768032 hasAuthorship W2988768032A5001910622 @default.
- W2988768032 hasAuthorship W2988768032A5053821067 @default.
- W2988768032 hasBestOaLocation W29887680322 @default.
- W2988768032 hasConcept C101738243 @default.
- W2988768032 hasConcept C108583219 @default.
- W2988768032 hasConcept C11413529 @default.
- W2988768032 hasConcept C147168706 @default.
- W2988768032 hasConcept C153180895 @default.
- W2988768032 hasConcept C154945302 @default.
- W2988768032 hasConcept C177973122 @default.
- W2988768032 hasConcept C202444582 @default.
- W2988768032 hasConcept C2776214188 @default.
- W2988768032 hasConcept C33923547 @default.
- W2988768032 hasConcept C38365724 @default.
- W2988768032 hasConcept C41008148 @default.
- W2988768032 hasConcept C50100734 @default.
- W2988768032 hasConcept C50644808 @default.
- W2988768032 hasConcept C81363708 @default.
- W2988768032 hasConcept C92757383 @default.
- W2988768032 hasConceptScore W2988768032C101738243 @default.
- W2988768032 hasConceptScore W2988768032C108583219 @default.
- W2988768032 hasConceptScore W2988768032C11413529 @default.
- W2988768032 hasConceptScore W2988768032C147168706 @default.
- W2988768032 hasConceptScore W2988768032C153180895 @default.
- W2988768032 hasConceptScore W2988768032C154945302 @default.
- W2988768032 hasConceptScore W2988768032C177973122 @default.
- W2988768032 hasConceptScore W2988768032C202444582 @default.
- W2988768032 hasConceptScore W2988768032C2776214188 @default.
- W2988768032 hasConceptScore W2988768032C33923547 @default.
- W2988768032 hasConceptScore W2988768032C38365724 @default.
- W2988768032 hasConceptScore W2988768032C41008148 @default.
- W2988768032 hasConceptScore W2988768032C50100734 @default.
- W2988768032 hasConceptScore W2988768032C50644808 @default.
- W2988768032 hasConceptScore W2988768032C81363708 @default.
- W2988768032 hasConceptScore W2988768032C92757383 @default.
- W2988768032 hasLocation W29887680321 @default.
- W2988768032 hasLocation W29887680322 @default.
- W2988768032 hasOpenAccess W2988768032 @default.
- W2988768032 hasPrimaryLocation W29887680321 @default.
- W2988768032 hasRelatedWork W2093503827 @default.
- W2988768032 hasRelatedWork W2148708551 @default.
- W2988768032 hasRelatedWork W2404774341 @default.
- W2988768032 hasRelatedWork W2750958979 @default.
- W2988768032 hasRelatedWork W2785668315 @default.
- W2988768032 hasRelatedWork W2883219711 @default.
- W2988768032 hasRelatedWork W2895596793 @default.
- W2988768032 hasRelatedWork W2949896761 @default.
- W2988768032 hasRelatedWork W2952051353 @default.
- W2988768032 hasRelatedWork W2964072166 @default.
- W2988768032 hasRelatedWork W2978219242 @default.
- W2988768032 hasRelatedWork W3005707914 @default.
- W2988768032 hasRelatedWork W3034581116 @default.
- W2988768032 hasRelatedWork W3035718760 @default.
- W2988768032 hasRelatedWork W3080338635 @default.
- W2988768032 hasRelatedWork W3089719735 @default.
- W2988768032 hasRelatedWork W3095615584 @default.
- W2988768032 hasRelatedWork W3098739061 @default.
- W2988768032 hasRelatedWork W3163875631 @default.
- W2988768032 hasRelatedWork W3183224688 @default.
- W2988768032 isParatext "false" @default.
- W2988768032 isRetracted "false" @default.
- W2988768032 magId "2988768032" @default.
- W2988768032 workType "article" @default.