Matches in SemOpenAlex for { <https://semopenalex.org/work/W2592451807> ?p ?o ?g. }
Showing items 1 to 55 of
55
with 100 items per page.
- W2592451807 endingPage "59" @default.
- W2592451807 startingPage "52" @default.
- W2592451807 abstract "While modern machine learning has made significant strides toward achieving high aptitude in various cognitive tasks, it does not provide any of the general abilities of mammalian intelligence. With a fundamentally different approach, Hierarchical Temporal Memory (HTM) is based on biological evidence that a common set of principles in the neocortex provides a diverse set of intelligent functions. Hierarchical temporal memories (HTMs) are biomimetic algorithms that can similarly be trained to perform inference and prediction on any temporal datastream. The Automata Processor (AP) is a configurable silicon implementation of nondeterministic finite automata, designed for massively parallel pattern matching. Key correspondences between counter-extended nondeterministic finite automata and the HTM activation model indicate use of the AP as an efficient hardware accelerator. In this article, the authors introduce a methodology for synthesizing HTMs on the Automata Processor, demonstrate three prediction applications on their model, and show its potential to achieve between 137 to 446 times speedup over the CPU." @default.
- W2592451807 created "2017-03-16" @default.
- W2592451807 creator A5010326266 @default.
- W2592451807 creator A5068766978 @default.
- W2592451807 creator A5083191451 @default.
- W2592451807 date "2017-01-01" @default.
- W2592451807 modified "2023-09-26" @default.
- W2592451807 title "Hierarchical Temporal Memory on the Automata Processor" @default.
- W2592451807 cites W2062949766 @default.
- W2592451807 doi "https://doi.org/10.1109/mm.2017.6" @default.
- W2592451807 hasPublicationYear "2017" @default.
- W2592451807 type Work @default.
- W2592451807 sameAs 2592451807 @default.
- W2592451807 citedByCount "9" @default.
- W2592451807 countsByYear W25924518072017 @default.
- W2592451807 countsByYear W25924518072018 @default.
- W2592451807 countsByYear W25924518072019 @default.
- W2592451807 countsByYear W25924518072020 @default.
- W2592451807 countsByYear W25924518072021 @default.
- W2592451807 countsByYear W25924518072022 @default.
- W2592451807 crossrefType "journal-article" @default.
- W2592451807 hasAuthorship W2592451807A5010326266 @default.
- W2592451807 hasAuthorship W2592451807A5068766978 @default.
- W2592451807 hasAuthorship W2592451807A5083191451 @default.
- W2592451807 hasConcept C112505250 @default.
- W2592451807 hasConcept C118524514 @default.
- W2592451807 hasConcept C173608175 @default.
- W2592451807 hasConcept C41008148 @default.
- W2592451807 hasConcept C80444323 @default.
- W2592451807 hasConceptScore W2592451807C112505250 @default.
- W2592451807 hasConceptScore W2592451807C118524514 @default.
- W2592451807 hasConceptScore W2592451807C173608175 @default.
- W2592451807 hasConceptScore W2592451807C41008148 @default.
- W2592451807 hasConceptScore W2592451807C80444323 @default.
- W2592451807 hasIssue "1" @default.
- W2592451807 hasLocation W25924518071 @default.
- W2592451807 hasOpenAccess W2592451807 @default.
- W2592451807 hasPrimaryLocation W25924518071 @default.
- W2592451807 hasRelatedWork W1491899005 @default.
- W2592451807 hasRelatedWork W1558545464 @default.
- W2592451807 hasRelatedWork W1604898313 @default.
- W2592451807 hasRelatedWork W1997145140 @default.
- W2592451807 hasRelatedWork W2034384303 @default.
- W2592451807 hasRelatedWork W2074301136 @default.
- W2592451807 hasRelatedWork W2117014006 @default.
- W2592451807 hasRelatedWork W2172791042 @default.
- W2592451807 hasRelatedWork W2372170743 @default.
- W2592451807 hasRelatedWork W4233815414 @default.
- W2592451807 hasVolume "37" @default.
- W2592451807 isParatext "false" @default.
- W2592451807 isRetracted "false" @default.
- W2592451807 magId "2592451807" @default.
- W2592451807 workType "article" @default.