Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386427720> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4386427720 abstract "Artificial Neural Networks (ANN) are habitually trained via the back-propagation (BP) algorithm. This approach has been extremely successful: Current models like GPT-3 have O(10 <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>11</sup> ) parameters, are trained on O(10 <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>11</sup> ) words and produce awe-inspiring results. However, there are good reasons to look for alternative training methods: With current algorithms and hardware constraints sometimes only half the available computing power is actually used. This is due to a complicated interplay between the size of the ANN, the available memory, throughput limitations of interconnects, the architecture of the network of computers, and the training algorithm. Training a model like the aforementioned GPT-3 takes months and costs millions. A different training paradigm, which could make clever use of specialized hardware, may train large ANNs more efficiently." @default.
- W4386427720 created "2023-09-05" @default.
- W4386427720 creator A5012892331 @default.
- W4386427720 creator A5035697987 @default.
- W4386427720 creator A5043781083 @default.
- W4386427720 creator A5043933499 @default.
- W4386427720 creator A5047474914 @default.
- W4386427720 creator A5048435361 @default.
- W4386427720 creator A5058259063 @default.
- W4386427720 creator A5068236230 @default.
- W4386427720 creator A5069072421 @default.
- W4386427720 creator A5091232404 @default.
- W4386427720 date "2023-06-26" @default.
- W4386427720 modified "2023-10-17" @default.
- W4386427720 title "Artificial Neural Network Training on an Optical Processor via Direct Feedback Alignment" @default.
- W4386427720 cites W2103869034 @default.
- W4386427720 doi "https://doi.org/10.1109/cleo/europe-eqec57999.2023.10231380" @default.
- W4386427720 hasPublicationYear "2023" @default.
- W4386427720 type Work @default.
- W4386427720 citedByCount "0" @default.
- W4386427720 crossrefType "proceedings-article" @default.
- W4386427720 hasAuthorship W4386427720A5012892331 @default.
- W4386427720 hasAuthorship W4386427720A5035697987 @default.
- W4386427720 hasAuthorship W4386427720A5043781083 @default.
- W4386427720 hasAuthorship W4386427720A5043933499 @default.
- W4386427720 hasAuthorship W4386427720A5047474914 @default.
- W4386427720 hasAuthorship W4386427720A5048435361 @default.
- W4386427720 hasAuthorship W4386427720A5058259063 @default.
- W4386427720 hasAuthorship W4386427720A5068236230 @default.
- W4386427720 hasAuthorship W4386427720A5069072421 @default.
- W4386427720 hasAuthorship W4386427720A5091232404 @default.
- W4386427720 hasConcept C111919701 @default.
- W4386427720 hasConcept C113775141 @default.
- W4386427720 hasConcept C119857082 @default.
- W4386427720 hasConcept C121332964 @default.
- W4386427720 hasConcept C153294291 @default.
- W4386427720 hasConcept C154945302 @default.
- W4386427720 hasConcept C155032097 @default.
- W4386427720 hasConcept C157764524 @default.
- W4386427720 hasConcept C2777211547 @default.
- W4386427720 hasConcept C41008148 @default.
- W4386427720 hasConcept C50644808 @default.
- W4386427720 hasConcept C51632099 @default.
- W4386427720 hasConcept C555944384 @default.
- W4386427720 hasConceptScore W4386427720C111919701 @default.
- W4386427720 hasConceptScore W4386427720C113775141 @default.
- W4386427720 hasConceptScore W4386427720C119857082 @default.
- W4386427720 hasConceptScore W4386427720C121332964 @default.
- W4386427720 hasConceptScore W4386427720C153294291 @default.
- W4386427720 hasConceptScore W4386427720C154945302 @default.
- W4386427720 hasConceptScore W4386427720C155032097 @default.
- W4386427720 hasConceptScore W4386427720C157764524 @default.
- W4386427720 hasConceptScore W4386427720C2777211547 @default.
- W4386427720 hasConceptScore W4386427720C41008148 @default.
- W4386427720 hasConceptScore W4386427720C50644808 @default.
- W4386427720 hasConceptScore W4386427720C51632099 @default.
- W4386427720 hasConceptScore W4386427720C555944384 @default.
- W4386427720 hasLocation W43864277201 @default.
- W4386427720 hasOpenAccess W4386427720 @default.
- W4386427720 hasPrimaryLocation W43864277201 @default.
- W4386427720 hasRelatedWork W1977274214 @default.
- W4386427720 hasRelatedWork W2012689841 @default.
- W4386427720 hasRelatedWork W2106973078 @default.
- W4386427720 hasRelatedWork W2130899929 @default.
- W4386427720 hasRelatedWork W2141967631 @default.
- W4386427720 hasRelatedWork W2151121162 @default.
- W4386427720 hasRelatedWork W2157746493 @default.
- W4386427720 hasRelatedWork W2894173309 @default.
- W4386427720 hasRelatedWork W2961085424 @default.
- W4386427720 hasRelatedWork W4206400463 @default.
- W4386427720 isParatext "false" @default.
- W4386427720 isRetracted "false" @default.
- W4386427720 workType "article" @default.