Matches in SemOpenAlex for { <https://semopenalex.org/work/W2530484110> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W2530484110 abstract "Deep Neural Networks (DNNs) are a set of powerful yet computationally complex learning mechanisms that are projected to dominate various artificial intelligence and massive data analytic domains. Physical viability, such as timing, memory, or energy efficiency, are standing challenges in realizing the true potential of DNNs. We propose DeLight, a set of novel methodologies which aim to bring physical constraints as design parameters in the training and execution of DNN architectures. We use physical profiling to bound the network size in accordance to the pertinent platform's characteristics. An automated customization methodology is proposed to adaptively conform the DNN configurations to meet the characterization of the underlying hardware while minimally affecting the inference accuracy. The key to our approach is a new content- and resource-aware transformation of data to a lower-dimensional embedding by which learning the correlation between data samples requires significantly smaller number of neurons. We leverage the performance gain achieved as a result of the data transformation to enable the training of multiple DNN architectures that can be aggregated to further boost the inference accuracy. An accompanying API is also developed, which can be used for rapid prototyping of an arbitrary DNN application customized to the platform. Proof-of concept evaluations for deployment of different imaging, audio, and smart-sensing applications demonstrate up to 100-fold performance improvement compared to the state-of-the-art DNN solutions." @default.
- W2530484110 created "2016-10-21" @default.
- W2530484110 creator A5019931011 @default.
- W2530484110 creator A5035073741 @default.
- W2530484110 creator A5070731184 @default.
- W2530484110 date "2016-10-01" @default.
- W2530484110 modified "2023-09-24" @default.
- W2530484110 title "Going deeper than deep learning for massive data analytics under physical constraints" @default.
- W2530484110 cites W2128260507 @default.
- W2530484110 cites W2407983593 @default.
- W2530484110 cites W2488255893 @default.
- W2530484110 doi "https://doi.org/10.1145/2968456.2976766" @default.
- W2530484110 hasPublicationYear "2016" @default.
- W2530484110 type Work @default.
- W2530484110 sameAs 2530484110 @default.
- W2530484110 citedByCount "2" @default.
- W2530484110 countsByYear W25304841102017 @default.
- W2530484110 countsByYear W25304841102021 @default.
- W2530484110 crossrefType "proceedings-article" @default.
- W2530484110 hasAuthorship W2530484110A5019931011 @default.
- W2530484110 hasAuthorship W2530484110A5035073741 @default.
- W2530484110 hasAuthorship W2530484110A5070731184 @default.
- W2530484110 hasBestOaLocation W25304841101 @default.
- W2530484110 hasConcept C107457646 @default.
- W2530484110 hasConcept C108583219 @default.
- W2530484110 hasConcept C124101348 @default.
- W2530484110 hasConcept C154945302 @default.
- W2530484110 hasConcept C175801342 @default.
- W2530484110 hasConcept C2522767166 @default.
- W2530484110 hasConcept C2777648619 @default.
- W2530484110 hasConcept C41008148 @default.
- W2530484110 hasConcept C79158427 @default.
- W2530484110 hasConceptScore W2530484110C107457646 @default.
- W2530484110 hasConceptScore W2530484110C108583219 @default.
- W2530484110 hasConceptScore W2530484110C124101348 @default.
- W2530484110 hasConceptScore W2530484110C154945302 @default.
- W2530484110 hasConceptScore W2530484110C175801342 @default.
- W2530484110 hasConceptScore W2530484110C2522767166 @default.
- W2530484110 hasConceptScore W2530484110C2777648619 @default.
- W2530484110 hasConceptScore W2530484110C41008148 @default.
- W2530484110 hasConceptScore W2530484110C79158427 @default.
- W2530484110 hasFunder F4320337345 @default.
- W2530484110 hasLocation W25304841101 @default.
- W2530484110 hasOpenAccess W2530484110 @default.
- W2530484110 hasPrimaryLocation W25304841101 @default.
- W2530484110 hasRelatedWork W2477220168 @default.
- W2530484110 hasRelatedWork W2489668743 @default.
- W2530484110 hasRelatedWork W2783293565 @default.
- W2530484110 hasRelatedWork W2906062308 @default.
- W2530484110 hasRelatedWork W2955609745 @default.
- W2530484110 hasRelatedWork W3038196548 @default.
- W2530484110 hasRelatedWork W3092101869 @default.
- W2530484110 hasRelatedWork W3125750166 @default.
- W2530484110 hasRelatedWork W3154404963 @default.
- W2530484110 hasRelatedWork W4255621315 @default.
- W2530484110 isParatext "false" @default.
- W2530484110 isRetracted "false" @default.
- W2530484110 magId "2530484110" @default.
- W2530484110 workType "article" @default.