Reconfigurable Stochastic Neurons Based on Strain Engineered Low Barrier Nanomagnets
| Title: | Reconfigurable Stochastic Neurons Based on Strain Engineered Low Barrier Nanomagnets |
|---|---|
| Authors: | Rahman, Rahnuma, Ganguly, Samiran, Bandyopadhyay, Supriyo |
| Publication Year: | 2024 |
| Collection: | Computer Science Condensed Matter |
| Subject Terms: | Computer Science - Emerging Technologies, Condensed Matter - Mesoscale and Nanoscale Physics, Electrical Engineering and Systems Science - Systems and Control |
| More Details: | Stochastic neurons are efficient hardware accelerators for solving a large variety of combinatorial optimization problems. "Binary" stochastic neurons (BSN) are those whose states fluctuate randomly between two levels +1 and -1, with the probability of being in either level determined by an external bias. "Analog" stochastic neurons (ASNs), in contrast, can assume any state between the two levels randomly (hence "analog") and can perform analog signal processing. They may be leveraged for such tasks as temporal sequence learning, processing and prediction. Both BSNs and ASNs can be used to build efficient and scalable neural networks. Both can be implemented with low (potential energy) barrier nanomagnets (LBMs) whose random magnetization orientations encode the binary or analog state variables. The difference between them is that the potential energy barrier in a BSN LBM, albeit low, is much higher than that in an ASN LBM. As a result, a BSN LBM has a clear double well potential profile, which makes its magnetization orientation assume one of two orientations at any time, resulting in the binary behavior. ASN nanomagnets, on the other hand, hardly have any energy barrier at all and hence lack the double well feature. That makes their magnetizations fluctuate in an analog fashion. Hence, one can reconfigure an ASN to a BSN, and vice-versa, by simply raising and lowering the energy barrier. If the LBM is magnetostrictive, then this can be done with local (electrically generated) strain. Such a reconfiguration capability heralds a powerful field programmable architecture for a p-computer, and the energy cost for this type of reconfiguration is miniscule. Comment: Some typos in the previous version have been corrected |
| Document Type: | Working Paper |
| Access URL: | http://arxiv.org/abs/2402.06168 |
| Accession Number: | edsarx.2402.06168 |
| Database: | arXiv |
| FullText | Text: Availability: 0 CustomLinks: – Url: http://arxiv.org/abs/2402.06168 Name: EDS - Arxiv Category: fullText Text: View this record from Arxiv MouseOverText: View this record from Arxiv – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edsarx&genre=article&issn=&ISBN=&volume=&issue=&date=20240208&spage=&pages=&title=Reconfigurable Stochastic Neurons Based on Strain Engineered Low Barrier Nanomagnets&atitle=Reconfigurable%20Stochastic%20Neurons%20Based%20on%20Strain%20Engineered%20Low%20Barrier%20Nanomagnets&aulast=Rahman%2C%20Rahnuma&id=DOI: Name: Full Text Finder (for New FTF UI) (s8985755) Category: fullText Text: Find It @ SCU Libraries MouseOverText: Find It @ SCU Libraries |
|---|---|
| Header | DbId: edsarx DbLabel: arXiv An: edsarx.2402.06168 RelevancyScore: 1085 AccessLevel: 3 PubType: Report PubTypeId: report PreciseRelevancyScore: 1085.39147949219 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Reconfigurable Stochastic Neurons Based on Strain Engineered Low Barrier Nanomagnets – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Rahman%2C+Rahnuma%22">Rahman, Rahnuma</searchLink><br /><searchLink fieldCode="AR" term="%22Ganguly%2C+Samiran%22">Ganguly, Samiran</searchLink><br /><searchLink fieldCode="AR" term="%22Bandyopadhyay%2C+Supriyo%22">Bandyopadhyay, Supriyo</searchLink> – Name: DatePubCY Label: Publication Year Group: Date Data: 2024 – Name: Subset Label: Collection Group: HoldingsInfo Data: Computer Science<br />Condensed Matter – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Computer+Science+-+Emerging+Technologies%22">Computer Science - Emerging Technologies</searchLink><br /><searchLink fieldCode="DE" term="%22Condensed+Matter+-+Mesoscale+and+Nanoscale+Physics%22">Condensed Matter - Mesoscale and Nanoscale Physics</searchLink><br /><searchLink fieldCode="DE" term="%22Electrical+Engineering+and+Systems+Science+-+Systems+and+Control%22">Electrical Engineering and Systems Science - Systems and Control</searchLink> – Name: Abstract Label: Description Group: Ab Data: Stochastic neurons are efficient hardware accelerators for solving a large variety of combinatorial optimization problems. "Binary" stochastic neurons (BSN) are those whose states fluctuate randomly between two levels +1 and -1, with the probability of being in either level determined by an external bias. "Analog" stochastic neurons (ASNs), in contrast, can assume any state between the two levels randomly (hence "analog") and can perform analog signal processing. They may be leveraged for such tasks as temporal sequence learning, processing and prediction. Both BSNs and ASNs can be used to build efficient and scalable neural networks. Both can be implemented with low (potential energy) barrier nanomagnets (LBMs) whose random magnetization orientations encode the binary or analog state variables. The difference between them is that the potential energy barrier in a BSN LBM, albeit low, is much higher than that in an ASN LBM. As a result, a BSN LBM has a clear double well potential profile, which makes its magnetization orientation assume one of two orientations at any time, resulting in the binary behavior. ASN nanomagnets, on the other hand, hardly have any energy barrier at all and hence lack the double well feature. That makes their magnetizations fluctuate in an analog fashion. Hence, one can reconfigure an ASN to a BSN, and vice-versa, by simply raising and lowering the energy barrier. If the LBM is magnetostrictive, then this can be done with local (electrically generated) strain. Such a reconfiguration capability heralds a powerful field programmable architecture for a p-computer, and the energy cost for this type of reconfiguration is miniscule.<br />Comment: Some typos in the previous version have been corrected – Name: TypeDocument Label: Document Type Group: TypDoc Data: Working Paper – Name: URL Label: Access URL Group: URL Data: <link linkTarget="URL" linkTerm="http://arxiv.org/abs/2402.06168" linkWindow="_blank">http://arxiv.org/abs/2402.06168</link> – Name: AN Label: Accession Number Group: ID Data: edsarx.2402.06168 |
| PLink | https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsarx&AN=edsarx.2402.06168 |
| RecordInfo | BibRecord: BibEntity: Subjects: – SubjectFull: Computer Science - Emerging Technologies Type: general – SubjectFull: Condensed Matter - Mesoscale and Nanoscale Physics Type: general – SubjectFull: Electrical Engineering and Systems Science - Systems and Control Type: general Titles: – TitleFull: Reconfigurable Stochastic Neurons Based on Strain Engineered Low Barrier Nanomagnets Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Rahman, Rahnuma – PersonEntity: Name: NameFull: Ganguly, Samiran – PersonEntity: Name: NameFull: Bandyopadhyay, Supriyo IsPartOfRelationships: – BibEntity: Dates: – D: 08 M: 02 Type: published Y: 2024 |
| ResultId | 1 |