Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287704904> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4287704904 abstract "Closing the gap between the hardware requirements of state-of-the-art convolutional neural networks and the limited resources constraining embedded applications is the next big challenge in deep learning research. The computational complexity and memory footprint of such neural networks are typically daunting for deployment in resource constrained environments. Model compression techniques, such as pruning, are emphasized among other optimization methods for solving this problem. Most existing techniques require domain expertise or result in irregular sparse representations, which increase the burden of deploying deep learning applications on embedded hardware accelerators. In this paper, we propose the autoencoder-based low-rank filter-sharing technique technique (ALF). When applied to various networks, ALF is compared to state-of-the-art pruning methods, demonstrating its efficient compression capabilities on theoretical metrics as well as on an accurate, deterministic hardware-model. In our experiments, ALF showed a reduction of 70% in network parameters, 61% in operations and 41% in execution time, with minimal loss in accuracy." @default.
- W4287704904 created "2022-07-26" @default.
- W4287704904 creator A5005732789 @default.
- W4287704904 creator A5037114730 @default.
- W4287704904 creator A5038840530 @default.
- W4287704904 creator A5043015774 @default.
- W4287704904 creator A5047063939 @default.
- W4287704904 creator A5079890456 @default.
- W4287704904 creator A5085220085 @default.
- W4287704904 date "2020-07-27" @default.
- W4287704904 modified "2023-09-29" @default.
- W4287704904 title "ALF: Autoencoder-based Low-rank Filter-sharing for Efficient Convolutional Neural Networks" @default.
- W4287704904 doi "https://doi.org/10.48550/arxiv.2007.13384" @default.
- W4287704904 hasPublicationYear "2020" @default.
- W4287704904 type Work @default.
- W4287704904 citedByCount "0" @default.
- W4287704904 crossrefType "posted-content" @default.
- W4287704904 hasAuthorship W4287704904A5005732789 @default.
- W4287704904 hasAuthorship W4287704904A5037114730 @default.
- W4287704904 hasAuthorship W4287704904A5038840530 @default.
- W4287704904 hasAuthorship W4287704904A5043015774 @default.
- W4287704904 hasAuthorship W4287704904A5047063939 @default.
- W4287704904 hasAuthorship W4287704904A5079890456 @default.
- W4287704904 hasAuthorship W4287704904A5085220085 @default.
- W4287704904 hasBestOaLocation W42877049041 @default.
- W4287704904 hasConcept C101738243 @default.
- W4287704904 hasConcept C106131492 @default.
- W4287704904 hasConcept C108010975 @default.
- W4287704904 hasConcept C108583219 @default.
- W4287704904 hasConcept C111335779 @default.
- W4287704904 hasConcept C111919701 @default.
- W4287704904 hasConcept C113775141 @default.
- W4287704904 hasConcept C114614502 @default.
- W4287704904 hasConcept C119857082 @default.
- W4287704904 hasConcept C154945302 @default.
- W4287704904 hasConcept C164226766 @default.
- W4287704904 hasConcept C2524010 @default.
- W4287704904 hasConcept C31972630 @default.
- W4287704904 hasConcept C33923547 @default.
- W4287704904 hasConcept C41008148 @default.
- W4287704904 hasConcept C50644808 @default.
- W4287704904 hasConcept C6557445 @default.
- W4287704904 hasConcept C74912251 @default.
- W4287704904 hasConcept C81363708 @default.
- W4287704904 hasConcept C86803240 @default.
- W4287704904 hasConceptScore W4287704904C101738243 @default.
- W4287704904 hasConceptScore W4287704904C106131492 @default.
- W4287704904 hasConceptScore W4287704904C108010975 @default.
- W4287704904 hasConceptScore W4287704904C108583219 @default.
- W4287704904 hasConceptScore W4287704904C111335779 @default.
- W4287704904 hasConceptScore W4287704904C111919701 @default.
- W4287704904 hasConceptScore W4287704904C113775141 @default.
- W4287704904 hasConceptScore W4287704904C114614502 @default.
- W4287704904 hasConceptScore W4287704904C119857082 @default.
- W4287704904 hasConceptScore W4287704904C154945302 @default.
- W4287704904 hasConceptScore W4287704904C164226766 @default.
- W4287704904 hasConceptScore W4287704904C2524010 @default.
- W4287704904 hasConceptScore W4287704904C31972630 @default.
- W4287704904 hasConceptScore W4287704904C33923547 @default.
- W4287704904 hasConceptScore W4287704904C41008148 @default.
- W4287704904 hasConceptScore W4287704904C50644808 @default.
- W4287704904 hasConceptScore W4287704904C6557445 @default.
- W4287704904 hasConceptScore W4287704904C74912251 @default.
- W4287704904 hasConceptScore W4287704904C81363708 @default.
- W4287704904 hasConceptScore W4287704904C86803240 @default.
- W4287704904 hasLocation W42877049041 @default.
- W4287704904 hasOpenAccess W4287704904 @default.
- W4287704904 hasPrimaryLocation W42877049041 @default.
- W4287704904 hasRelatedWork W2337926734 @default.
- W4287704904 hasRelatedWork W2732415564 @default.
- W4287704904 hasRelatedWork W3021430260 @default.
- W4287704904 hasRelatedWork W3044458868 @default.
- W4287704904 hasRelatedWork W3045422318 @default.
- W4287704904 hasRelatedWork W3092479081 @default.
- W4287704904 hasRelatedWork W4213225422 @default.
- W4287704904 hasRelatedWork W4287704904 @default.
- W4287704904 hasRelatedWork W4306194456 @default.
- W4287704904 hasRelatedWork W4311257506 @default.
- W4287704904 isParatext "false" @default.
- W4287704904 isRetracted "false" @default.
- W4287704904 workType "article" @default.