Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226284328> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4226284328 abstract "In recent years, a considerable research effort has shown the energy benefits of implementing neural networks with memristors or other emerging memory technologies. However, for extreme-edge applications with high uncertainty, access to reduced amounts of data, and where explainable decisions are required, neural networks may not provide an acceptable form of intelligence. Bayesian reasoning can solve these concerns, but it is computationally expensive and, unlike neural networks, does not translate naturally to memristor-based architectures. In this work, we introduce, demonstrate experimentally on a fully fabricated hybrid CMOS-memristor system, and analyze a Bayesian machine designed for highly-energy efficient Bayesian reasoning. The architecture of the machine is obtained by writing Bayes' law in a way making its implementation natural by the principles of distributed memory and stochastic computing, allowing the circuit to function using solely local memory and minimal data movement. Measurements on a fabricated small-scale Bayesian machine featuring 2,048 memristors and 30,080 transistors show the viability of this approach and the possibility of overcoming the challenges associated with its design: the inherent imperfections of memristors, as well as the need to distribute very locally higher-than-nominal supply voltages. The design of a scaled-up version of the machine shows its outstanding energy efficiency on a real-life gesture recognition task: a gesture can be recognized using 5,000 times less energy than using a microcontroller unit. The Bayesian machine also features several desirable features, e.g., instant on/off operation, compatibility with low supply voltages, and resilience to single-event upsets. These results open the road for Bayesian reasoning as an attractive way for energy-efficient, robust, and explainable intelligence at the edge." @default.
- W4226284328 created "2022-05-05" @default.
- W4226284328 creator A5018125818 @default.
- W4226284328 creator A5024303339 @default.
- W4226284328 creator A5031610889 @default.
- W4226284328 creator A5032162752 @default.
- W4226284328 creator A5046458712 @default.
- W4226284328 creator A5046559880 @default.
- W4226284328 creator A5053141872 @default.
- W4226284328 creator A5063819347 @default.
- W4226284328 creator A5068164263 @default.
- W4226284328 creator A5087494328 @default.
- W4226284328 date "2021-12-20" @default.
- W4226284328 modified "2023-09-23" @default.
- W4226284328 title "A Memristor-Based Bayesian Machine" @default.
- W4226284328 doi "https://doi.org/10.48550/arxiv.2112.10547" @default.
- W4226284328 hasPublicationYear "2021" @default.
- W4226284328 type Work @default.
- W4226284328 citedByCount "0" @default.
- W4226284328 crossrefType "posted-content" @default.
- W4226284328 hasAuthorship W4226284328A5018125818 @default.
- W4226284328 hasAuthorship W4226284328A5024303339 @default.
- W4226284328 hasAuthorship W4226284328A5031610889 @default.
- W4226284328 hasAuthorship W4226284328A5032162752 @default.
- W4226284328 hasAuthorship W4226284328A5046458712 @default.
- W4226284328 hasAuthorship W4226284328A5046559880 @default.
- W4226284328 hasAuthorship W4226284328A5053141872 @default.
- W4226284328 hasAuthorship W4226284328A5063819347 @default.
- W4226284328 hasAuthorship W4226284328A5068164263 @default.
- W4226284328 hasAuthorship W4226284328A5087494328 @default.
- W4226284328 hasBestOaLocation W42262843281 @default.
- W4226284328 hasConcept C107673813 @default.
- W4226284328 hasConcept C119857082 @default.
- W4226284328 hasConcept C127413603 @default.
- W4226284328 hasConcept C149635348 @default.
- W4226284328 hasConcept C150072547 @default.
- W4226284328 hasConcept C154945302 @default.
- W4226284328 hasConcept C173018170 @default.
- W4226284328 hasConcept C24326235 @default.
- W4226284328 hasConcept C41008148 @default.
- W4226284328 hasConcept C50644808 @default.
- W4226284328 hasConceptScore W4226284328C107673813 @default.
- W4226284328 hasConceptScore W4226284328C119857082 @default.
- W4226284328 hasConceptScore W4226284328C127413603 @default.
- W4226284328 hasConceptScore W4226284328C149635348 @default.
- W4226284328 hasConceptScore W4226284328C150072547 @default.
- W4226284328 hasConceptScore W4226284328C154945302 @default.
- W4226284328 hasConceptScore W4226284328C173018170 @default.
- W4226284328 hasConceptScore W4226284328C24326235 @default.
- W4226284328 hasConceptScore W4226284328C41008148 @default.
- W4226284328 hasConceptScore W4226284328C50644808 @default.
- W4226284328 hasLocation W42262843281 @default.
- W4226284328 hasOpenAccess W4226284328 @default.
- W4226284328 hasPrimaryLocation W42262843281 @default.
- W4226284328 hasRelatedWork W2019421138 @default.
- W4226284328 hasRelatedWork W2767624311 @default.
- W4226284328 hasRelatedWork W2895726985 @default.
- W4226284328 hasRelatedWork W2961085424 @default.
- W4226284328 hasRelatedWork W3215828240 @default.
- W4226284328 hasRelatedWork W4280603446 @default.
- W4226284328 hasRelatedWork W4286629047 @default.
- W4226284328 hasRelatedWork W4306674287 @default.
- W4226284328 hasRelatedWork W1629725936 @default.
- W4226284328 hasRelatedWork W4224009465 @default.
- W4226284328 isParatext "false" @default.
- W4226284328 isRetracted "false" @default.
- W4226284328 workType "article" @default.