dc.contributor.author | Filippas, Dionysios | en |
dc.contributor.author | Nicopoulos, Chrysostomos | en |
dc.contributor.author | Dimitrakopoulos, Giorgos | en |
dc.contributor.editor | Mircea, Stan | en |
dc.contributor.editor | Xinfei, Guo | en |
dc.creator | Filippas, Dionysios | en |
dc.creator | Nicopoulos, Chrysostomos | en |
dc.creator | Dimitrakopoulos, Giorgos | en |
dc.date.accessioned | 2023-12-19T15:13:22Z | |
dc.date.available | 2023-12-19T15:13:22Z | |
dc.date.issued | 2023-01-09 | |
dc.identifier.issn | 1557-9999 | |
dc.identifier.uri | http://gnosis.library.ucy.ac.cy/handle/7/65819 | en |
dc.description.abstract | Convolution is one of the most critical operations in various application domains and its computation should combine high performance with energy efficiency. This requirement is critical both for standard convolution and for its other spatial variants, such as dilated, strided, or transposed convolutions. In this work, we focus on the design of a streaming convolution engine, called LazyDCstream, that is tuned for dilated convolution. LazyDCstream utilizes a sliding-window architecture for input data reuse and leverages the already-known decomposition of dilated convolution to: (a) maximize window buffer sharing and (b) enable “lazy” data movement that keeps data transfers per clock cycle as few as possible, and, most importantly, independent of the dilation rate. These two architectural features reduce the power consumption relative to efficient streaming convolution engines without introducing any throughput or area penalty. | en |
dc.language.iso | eng | en |
dc.publisher | IEEE | en |
dc.source | IEEE Transactions on Very Large Scale Integration (VLSI) Systems | en |
dc.source.uri | https://ieeexplore.ieee.org/abstract/document/10012295 | en |
dc.subject | Convolutional neural network | en |
dc.subject | Dilated convolution | en |
dc.subject | Low-power design | en |
dc.subject | Spatial filtering | en |
dc.title | Streaming Dilated Convolution Engine | en |
dc.type | info:eu-repo/semantics/article | en |
dc.identifier.doi | 10.1109/TVLSI.2022.3233882 | en |
dc.description.volume | 31 | |
dc.description.issue | 3 | |
dc.description.startingpage | 401 | |
dc.description.endingpage | 405 | |
dc.author.faculty | 007 Πολυτεχνική Σχολή / Faculty of Engineering | |
dc.author.department | Τμήμα Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών / Department of Electrical and Computer Engineering | |
dc.type.uhtype | Article | en |
dc.contributor.orcid | Filippas, Dionysios [0000-0002-4729-3336] | |
dc.contributor.orcid | Nicopoulos, Chrysostomos [0000-0001-6389-6068] | |
dc.contributor.orcid | Dimitrakopoulos, Giorgos [0000-0003-3688-7865] | |
dc.type.subtype | SCIENTIFIC_JOURNAL | en |
dc.gnosis.orcid | 0000-0002-4729-3336 | |
dc.gnosis.orcid | 0000-0001-6389-6068 | |
dc.gnosis.orcid | 0000-0003-3688-7865 | |