Matches in SemOpenAlex for { <https://semopenalex.org/work/W2742437467> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2742437467 abstract "Artificial neural networks are a popular and effective machine learning technique. Great progress has been made parallelizing the expensive training phase of an individual network, leading to highly specialized pieces of hardware, many based on GPU-type architectures, and more concurrent algorithms such as synthetic gradients. However, the training phase continues to be a bottleneck, where the training data must be processed serially over thousands of individual training runs. This work considers a multigrid reduction in time (MGRIT) algorithm that is able to parallelize over the thousands of training runs and converge to the exact same solution as traditional training would provide. MGRIT was originally developed to provide parallelism for time evolution problems that serially step through a finite number of time-steps. This work recasts the training of a neural network similarly, treating neural network training as an evolution equation that evolves the network weights from one step to the next. Thus, this work concerns distributed computing approaches for neural networks, but is distinct from other approaches which seek to parallelize only over individual training runs. The work concludes with supporting numerical results for two model problems." @default.
- W2742437467 created "2017-08-17" @default.
- W2742437467 creator A5038235793 @default.
- W2742437467 date "2017-08-07" @default.
- W2742437467 modified "2023-09-27" @default.
- W2742437467 title "Parallelizing Over Artificial Neural Network Training Runs with Multigrid" @default.
- W2742437467 cites W2151317657 @default.
- W2742437467 cites W2178031510 @default.
- W2742437467 cites W2765729611 @default.
- W2742437467 hasPublicationYear "2017" @default.
- W2742437467 type Work @default.
- W2742437467 sameAs 2742437467 @default.
- W2742437467 citedByCount "3" @default.
- W2742437467 countsByYear W27424374672018 @default.
- W2742437467 crossrefType "posted-content" @default.
- W2742437467 hasAuthorship W2742437467A5038235793 @default.
- W2742437467 hasConcept C111335779 @default.
- W2742437467 hasConcept C119857082 @default.
- W2742437467 hasConcept C121332964 @default.
- W2742437467 hasConcept C137119250 @default.
- W2742437467 hasConcept C149635348 @default.
- W2742437467 hasConcept C153294291 @default.
- W2742437467 hasConcept C154945302 @default.
- W2742437467 hasConcept C158622935 @default.
- W2742437467 hasConcept C173608175 @default.
- W2742437467 hasConcept C2524010 @default.
- W2742437467 hasConcept C2777211547 @default.
- W2742437467 hasConcept C2780513914 @default.
- W2742437467 hasConcept C2781172179 @default.
- W2742437467 hasConcept C33923547 @default.
- W2742437467 hasConcept C41008148 @default.
- W2742437467 hasConcept C50644808 @default.
- W2742437467 hasConcept C62520636 @default.
- W2742437467 hasConceptScore W2742437467C111335779 @default.
- W2742437467 hasConceptScore W2742437467C119857082 @default.
- W2742437467 hasConceptScore W2742437467C121332964 @default.
- W2742437467 hasConceptScore W2742437467C137119250 @default.
- W2742437467 hasConceptScore W2742437467C149635348 @default.
- W2742437467 hasConceptScore W2742437467C153294291 @default.
- W2742437467 hasConceptScore W2742437467C154945302 @default.
- W2742437467 hasConceptScore W2742437467C158622935 @default.
- W2742437467 hasConceptScore W2742437467C173608175 @default.
- W2742437467 hasConceptScore W2742437467C2524010 @default.
- W2742437467 hasConceptScore W2742437467C2777211547 @default.
- W2742437467 hasConceptScore W2742437467C2780513914 @default.
- W2742437467 hasConceptScore W2742437467C2781172179 @default.
- W2742437467 hasConceptScore W2742437467C33923547 @default.
- W2742437467 hasConceptScore W2742437467C41008148 @default.
- W2742437467 hasConceptScore W2742437467C50644808 @default.
- W2742437467 hasConceptScore W2742437467C62520636 @default.
- W2742437467 hasLocation W27424374671 @default.
- W2742437467 hasOpenAccess W2742437467 @default.
- W2742437467 hasPrimaryLocation W27424374671 @default.
- W2742437467 hasRelatedWork W1809453380 @default.
- W2742437467 hasRelatedWork W2023366022 @default.
- W2742437467 hasRelatedWork W2105872130 @default.
- W2742437467 hasRelatedWork W2182420680 @default.
- W2742437467 hasRelatedWork W2248968530 @default.
- W2742437467 hasRelatedWork W2320747856 @default.
- W2742437467 hasRelatedWork W2768571798 @default.
- W2742437467 hasRelatedWork W2768683495 @default.
- W2742437467 hasRelatedWork W2807522660 @default.
- W2742437467 hasRelatedWork W2952926545 @default.
- W2742437467 hasRelatedWork W2991165082 @default.
- W2742437467 hasRelatedWork W3020408629 @default.
- W2742437467 hasRelatedWork W3044664263 @default.
- W2742437467 hasRelatedWork W3080524058 @default.
- W2742437467 hasRelatedWork W3086105743 @default.
- W2742437467 hasRelatedWork W3125673699 @default.
- W2742437467 hasRelatedWork W3178283644 @default.
- W2742437467 hasRelatedWork W3188325035 @default.
- W2742437467 hasRelatedWork W3190161200 @default.
- W2742437467 hasRelatedWork W3191116967 @default.
- W2742437467 isParatext "false" @default.
- W2742437467 isRetracted "false" @default.
- W2742437467 magId "2742437467" @default.
- W2742437467 workType "article" @default.