Matches in SemOpenAlex for { <https://semopenalex.org/work/W1594170634> ?p ?o ?g. }
- W1594170634 abstract "We train a set of state of the art neural networks, the Maxout networks (Goodfellow et al., 2013a), on three benchmark datasets: the MNIST, CIFAR10 and SVHN, with three distinct storing formats: floating point, fixed point and dynamic fixed point. For each of those datasets and for each of those formats, we assess the impact of the precision of the storage on the final error of the training. We find that very low precision storage is sufficient not just for running trained networks but also for training them. For example, Maxout networks state-of-the-art results are nearly maintained with 10 bits for storing activations and gradients, and 12 bits for storing parameters." @default.
- W1594170634 created "2016-06-24" @default.
- W1594170634 creator A5039449234 @default.
- W1594170634 creator A5067086966 @default.
- W1594170634 creator A5086198262 @default.
- W1594170634 date "2014-12-22" @default.
- W1594170634 modified "2023-09-27" @default.
- W1594170634 title "Low precision storage for deep learning" @default.
- W1594170634 cites W1492459858 @default.
- W1594170634 cites W1720204209 @default.
- W1594170634 cites W1841592590 @default.
- W1594170634 cites W1872489089 @default.
- W1594170634 cites W1904365287 @default.
- W1594170634 cites W1907282891 @default.
- W1594170634 cites W2048266589 @default.
- W1594170634 cites W2074488431 @default.
- W1594170634 cites W2112796928 @default.
- W1594170634 cites W2122221396 @default.
- W1594170634 cites W2130017646 @default.
- W1594170634 cites W2149428536 @default.
- W1594170634 cites W2153412052 @default.
- W1594170634 cites W2153846939 @default.
- W1594170634 cites W2168231600 @default.
- W1594170634 cites W2168839239 @default.
- W1594170634 cites W2335728318 @default.
- W1594170634 cites W2546302380 @default.
- W1594170634 cites W2552956953 @default.
- W1594170634 cites W2949888546 @default.
- W1594170634 cites W3037950864 @default.
- W1594170634 cites W3118608800 @default.
- W1594170634 cites W587794757 @default.
- W1594170634 hasPublicationYear "2014" @default.
- W1594170634 type Work @default.
- W1594170634 sameAs 1594170634 @default.
- W1594170634 citedByCount "112" @default.
- W1594170634 countsByYear W15941706342015 @default.
- W1594170634 countsByYear W15941706342016 @default.
- W1594170634 countsByYear W15941706342017 @default.
- W1594170634 countsByYear W15941706342018 @default.
- W1594170634 countsByYear W15941706342019 @default.
- W1594170634 countsByYear W15941706342020 @default.
- W1594170634 countsByYear W15941706342021 @default.
- W1594170634 countsByYear W15941706342022 @default.
- W1594170634 crossrefType "posted-content" @default.
- W1594170634 hasAuthorship W1594170634A5039449234 @default.
- W1594170634 hasAuthorship W1594170634A5067086966 @default.
- W1594170634 hasAuthorship W1594170634A5086198262 @default.
- W1594170634 hasConcept C108583219 @default.
- W1594170634 hasConcept C11413529 @default.
- W1594170634 hasConcept C119857082 @default.
- W1594170634 hasConcept C13280743 @default.
- W1594170634 hasConcept C154945302 @default.
- W1594170634 hasConcept C177264268 @default.
- W1594170634 hasConcept C185798385 @default.
- W1594170634 hasConcept C190502265 @default.
- W1594170634 hasConcept C199360897 @default.
- W1594170634 hasConcept C205649164 @default.
- W1594170634 hasConcept C2524010 @default.
- W1594170634 hasConcept C28719098 @default.
- W1594170634 hasConcept C2984842247 @default.
- W1594170634 hasConcept C33923547 @default.
- W1594170634 hasConcept C41008148 @default.
- W1594170634 hasConcept C48103436 @default.
- W1594170634 hasConcept C50644808 @default.
- W1594170634 hasConcept C51632099 @default.
- W1594170634 hasConceptScore W1594170634C108583219 @default.
- W1594170634 hasConceptScore W1594170634C11413529 @default.
- W1594170634 hasConceptScore W1594170634C119857082 @default.
- W1594170634 hasConceptScore W1594170634C13280743 @default.
- W1594170634 hasConceptScore W1594170634C154945302 @default.
- W1594170634 hasConceptScore W1594170634C177264268 @default.
- W1594170634 hasConceptScore W1594170634C185798385 @default.
- W1594170634 hasConceptScore W1594170634C190502265 @default.
- W1594170634 hasConceptScore W1594170634C199360897 @default.
- W1594170634 hasConceptScore W1594170634C205649164 @default.
- W1594170634 hasConceptScore W1594170634C2524010 @default.
- W1594170634 hasConceptScore W1594170634C28719098 @default.
- W1594170634 hasConceptScore W1594170634C2984842247 @default.
- W1594170634 hasConceptScore W1594170634C33923547 @default.
- W1594170634 hasConceptScore W1594170634C41008148 @default.
- W1594170634 hasConceptScore W1594170634C48103436 @default.
- W1594170634 hasConceptScore W1594170634C50644808 @default.
- W1594170634 hasConceptScore W1594170634C51632099 @default.
- W1594170634 hasLocation W15941706341 @default.
- W1594170634 hasOpenAccess W1594170634 @default.
- W1594170634 hasPrimaryLocation W15941706341 @default.
- W1594170634 hasRelatedWork W1522301498 @default.
- W1594170634 hasRelatedWork W1686810756 @default.
- W1594170634 hasRelatedWork W1724438581 @default.
- W1594170634 hasRelatedWork W1821462560 @default.
- W1594170634 hasRelatedWork W1825672851 @default.
- W1594170634 hasRelatedWork W1841592590 @default.
- W1594170634 hasRelatedWork W1902934009 @default.
- W1594170634 hasRelatedWork W2108598243 @default.
- W1594170634 hasRelatedWork W2112796928 @default.
- W1594170634 hasRelatedWork W2194775991 @default.
- W1594170634 hasRelatedWork W2279098554 @default.
- W1594170634 hasRelatedWork W2300242332 @default.
- W1594170634 hasRelatedWork W2319920447 @default.
- W1594170634 hasRelatedWork W2469490737 @default.