Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100813974> ?p ?o ?g. }
- W3100813974 endingPage "50" @default.
- W3100813974 startingPage "40" @default.
- W3100813974 abstract "Recent breakthroughs in Deep Neural Networks (DNNs) have fueled a growing demand for domain-specific hardware accelerators (i.e., DNN chips). However, designing DNN chips is non-trivial because: (1) mainstream DNNs have millions of parameters and billions of operations; (2) the design space is large due to numerous design choices of dataflows, processing elements, memory hierarchy, etc.; and (3) there is an algorithm/hardware co-design need for the same DNN functionality to have a different decomposition that would require different hardware IPs and thus correspond to dramatically different performance/energy/area tradeoffs. Therefore, DNN chips often take months to years to design and require a large team of cross-disciplinary experts. To enable fast and effective DNN chip design, we propose AutoDNNchip - a DNN chip generator that can automatically produce both FPGA- and ASIC-based DNN chip implementation (i.e., synthesizable RTL code with optimized algorithm-to-hardware mapping) from DNNs developed by machine learning frameworks (e.g., PyTorch) for a designated application and dataset without humans in the loop. Specifically, AutoDNNchip consists of 2 integrated enablers: (1) a Chip Predictor, which can accurately and efficiently predict a DNN accelerator's energy, throughput, latency, and area based on the DNN model parameters, hardware configurations, technology-based IPs, and platform constraints; and (2) a Chip Builder, which can automatically explore the design space of DNN chips (including IP selections, block configurations, resource balancing, etc.), optimize chip designs via the Chip Predictor, and then generate synthesizable RTL code with optimized dataflows to achieve the target design metrics. Experimental results show that our Chip Predictor's predicted performance differs from real-measured ones by" @default.
- W3100813974 created "2020-11-23" @default.
- W3100813974 creator A5000680423 @default.
- W3100813974 creator A5008678513 @default.
- W3100813974 creator A5019582323 @default.
- W3100813974 creator A5025414128 @default.
- W3100813974 creator A5026072485 @default.
- W3100813974 creator A5053022655 @default.
- W3100813974 creator A5056321228 @default.
- W3100813974 creator A5060183082 @default.
- W3100813974 creator A5065172226 @default.
- W3100813974 creator A5075140744 @default.
- W3100813974 date "2020-02-23" @default.
- W3100813974 modified "2023-09-28" @default.
- W3100813974 title "AutoDNNchip: An Automated DNN Chip Predictor and Builder for Both FPGAs and ASICs" @default.
- W3100813974 cites W1982052956 @default.
- W3100813974 cites W2094756095 @default.
- W3100813974 cites W2108598243 @default.
- W3100813974 cites W2137966900 @default.
- W3100813974 cites W2194775991 @default.
- W3100813974 cites W2289252105 @default.
- W3100813974 cites W2396572963 @default.
- W3100813974 cites W2519224033 @default.
- W3100813974 cites W2566876368 @default.
- W3100813974 cites W2606722458 @default.
- W3100813974 cites W2727238169 @default.
- W3100813974 cites W2759777444 @default.
- W3100813974 cites W2762910930 @default.
- W3100813974 cites W2802331610 @default.
- W3100813974 cites W2860338957 @default.
- W3100813974 cites W2887782013 @default.
- W3100813974 cites W2899915146 @default.
- W3100813974 cites W2940862705 @default.
- W3100813974 cites W2962856739 @default.
- W3100813974 cites W2963163009 @default.
- W3100813974 cites W2964010167 @default.
- W3100813974 cites W2997929983 @default.
- W3100813974 cites W3037913581 @default.
- W3100813974 hasPublicationYear "2020" @default.
- W3100813974 type Work @default.
- W3100813974 sameAs 3100813974 @default.
- W3100813974 citedByCount "17" @default.
- W3100813974 countsByYear W31008139742019 @default.
- W3100813974 countsByYear W31008139742020 @default.
- W3100813974 countsByYear W31008139742021 @default.
- W3100813974 crossrefType "proceedings-article" @default.
- W3100813974 hasAuthorship W3100813974A5000680423 @default.
- W3100813974 hasAuthorship W3100813974A5008678513 @default.
- W3100813974 hasAuthorship W3100813974A5019582323 @default.
- W3100813974 hasAuthorship W3100813974A5025414128 @default.
- W3100813974 hasAuthorship W3100813974A5026072485 @default.
- W3100813974 hasAuthorship W3100813974A5053022655 @default.
- W3100813974 hasAuthorship W3100813974A5056321228 @default.
- W3100813974 hasAuthorship W3100813974A5060183082 @default.
- W3100813974 hasAuthorship W3100813974A5065172226 @default.
- W3100813974 hasAuthorship W3100813974A5075140744 @default.
- W3100813974 hasBestOaLocation W31008139741 @default.
- W3100813974 hasConcept C111919701 @default.
- W3100813974 hasConcept C118021083 @default.
- W3100813974 hasConcept C118524514 @default.
- W3100813974 hasConcept C149635348 @default.
- W3100813974 hasConcept C157764524 @default.
- W3100813974 hasConcept C165005293 @default.
- W3100813974 hasConcept C2776221188 @default.
- W3100813974 hasConcept C41008148 @default.
- W3100813974 hasConcept C42935608 @default.
- W3100813974 hasConcept C555944384 @default.
- W3100813974 hasConcept C76155785 @default.
- W3100813974 hasConcept C77390884 @default.
- W3100813974 hasConcept C9390403 @default.
- W3100813974 hasConceptScore W3100813974C111919701 @default.
- W3100813974 hasConceptScore W3100813974C118021083 @default.
- W3100813974 hasConceptScore W3100813974C118524514 @default.
- W3100813974 hasConceptScore W3100813974C149635348 @default.
- W3100813974 hasConceptScore W3100813974C157764524 @default.
- W3100813974 hasConceptScore W3100813974C165005293 @default.
- W3100813974 hasConceptScore W3100813974C2776221188 @default.
- W3100813974 hasConceptScore W3100813974C41008148 @default.
- W3100813974 hasConceptScore W3100813974C42935608 @default.
- W3100813974 hasConceptScore W3100813974C555944384 @default.
- W3100813974 hasConceptScore W3100813974C76155785 @default.
- W3100813974 hasConceptScore W3100813974C77390884 @default.
- W3100813974 hasConceptScore W3100813974C9390403 @default.
- W3100813974 hasLocation W31008139741 @default.
- W3100813974 hasLocation W31008139742 @default.
- W3100813974 hasOpenAccess W3100813974 @default.
- W3100813974 hasPrimaryLocation W31008139741 @default.
- W3100813974 hasRelatedWork W1686810756 @default.
- W3100813974 hasRelatedWork W2067523571 @default.
- W3100813974 hasRelatedWork W2094756095 @default.
- W3100813974 hasRelatedWork W2152839228 @default.
- W3100813974 hasRelatedWork W2194775991 @default.
- W3100813974 hasRelatedWork W2276486856 @default.
- W3100813974 hasRelatedWork W2279098554 @default.
- W3100813974 hasRelatedWork W2285660444 @default.
- W3100813974 hasRelatedWork W2289252105 @default.
- W3100813974 hasRelatedWork W2565851976 @default.
- W3100813974 hasRelatedWork W2606722458 @default.