000906936 001__ 906936
000906936 005__ 20230328130144.0
000906936 0247_ $$2doi$$a10.1002/aelm.202101198
000906936 0247_ $$2Handle$$a2128/33995
000906936 0247_ $$2WOS$$aWOS:000762807700001
000906936 037__ $$aFZJ-2022-01761
000906936 082__ $$a621.3
000906936 1001_ $$0P:(DE-HGF)0$$aPedretti, Giacomo$$b0
000906936 245__ $$aDifferentiable Content Addressable Memory with Memristors
000906936 260__ $$aWeinheim$$bWiley-VCH Verlag GmbH & Co. KG$$c2022
000906936 3367_ $$2DRIVER$$aarticle
000906936 3367_ $$2DataCite$$aOutput Types/Journal article
000906936 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1677491013_26873
000906936 3367_ $$2BibTeX$$aARTICLE
000906936 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000906936 3367_ $$00$$2EndNote$$aJournal Article
000906936 520__ $$aMemristors, Flash, and related nonvolatile analog device technologies offer in-memory computing structures operating in the analog domain, such as accelerating linear matrix operations in array structures. These take advantage of analog tunability and large dynamic range. At the other side, content addressable memories (CAM) are fast digital lookup tables which effectively perform nonlinear Boolean logic and return a digital match/mismatch value. Recently, nonvolatile analog CAMs have been presented merging analog storage and analog search operations with digital match/mismatch output. However, CAM blocks cannot easily be inserted within a larger adaptive system due to the challenges of training and learning with binary outputs. Here, a missing link between analog crossbar arrays and CAMs, namely a differentiable content addressable memory (dCAM), is presented. Utilizing nonvolatile memories that act as a “soft” memory with analog outputs, dCAM enables learning and fine-tuning of the memory operation and performance. Four applications are quantitatively evaluated to highlight the capabilities: improved data pattern storage, improved robustness to noise and variability, reduced energy and latency performance, and an application to solving Boolean satisfiability optimization problems. The use of dCAM is envisioned as a core building block of fully differentiable computing systems employing multiple types of analog compute operations and memories.
000906936 536__ $$0G:(DE-HGF)POF4-5234$$a5234 - Emerging NC Architectures (POF4-523)$$cPOF4-523$$fPOF IV$$x0
000906936 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
000906936 7001_ $$0P:(DE-HGF)0$$aGraves, Catherine E.$$b1
000906936 7001_ $$0P:(DE-HGF)0$$aVan Vaerenbergh, Thomas$$b2
000906936 7001_ $$0P:(DE-HGF)0$$aSerebryakov, Sergey$$b3
000906936 7001_ $$0P:(DE-HGF)0$$aFoltin, Martin$$b4
000906936 7001_ $$0P:(DE-HGF)0$$aSheng, Xia$$b5
000906936 7001_ $$0P:(DE-HGF)0$$aMao, Ruibin$$b6
000906936 7001_ $$0P:(DE-HGF)0$$aLi, Can$$b7
000906936 7001_ $$0P:(DE-Juel1)188145$$aStrachan, John Paul$$b8$$eCorresponding author
000906936 773__ $$0PERI:(DE-600)2810904-1$$a10.1002/aelm.202101198$$gp. 2101198 -$$n8$$p2101198 -$$tAdvanced electronic materials$$v8$$x2199-160X$$y2022
000906936 8564_ $$uhttps://juser.fz-juelich.de/record/906936/files/Adv%20Elect%20Materials%20-%202022%20-%20Pedretti%20-%20Differentiable%20Content%20Addressable%20Memory%20with%20Memristors.pdf$$yOpenAccess
000906936 8767_ $$d2022-02-16$$eHybrid-OA$$jDEAL
000906936 909CO $$ooai:juser.fz-juelich.de:906936$$pdnbdelivery$$popenCost$$pVDB$$pdriver$$pOpenAPC_DEAL$$popen_access$$popenaire
000906936 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188145$$aForschungszentrum Jülich$$b8$$kFZJ
000906936 9131_ $$0G:(DE-HGF)POF4-523$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5234$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vNeuromorphic Computing and Network Dynamics$$x0
000906936 9141_ $$y2022
000906936 915pc $$0PC:(DE-HGF)0000$$2APC$$aAPC keys set
000906936 915pc $$0PC:(DE-HGF)0001$$2APC$$aLocal Funding
000906936 915pc $$0PC:(DE-HGF)0002$$2APC$$aDFG OA Publikationskosten
000906936 915pc $$0PC:(DE-HGF)0120$$2APC$$aDEAL: Wiley 2019
000906936 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2021-01-28
000906936 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2021-01-28
000906936 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bADV ELECTRON MATER : 2019$$d2021-01-28
000906936 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bADV ELECTRON MATER : 2019$$d2021-01-28
000906936 915__ $$0StatID:(DE-HGF)3001$$2StatID$$aDEAL Wiley$$d2021-01-28$$wger
000906936 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2021-01-28
000906936 915__ $$0LIC:(DE-HGF)CCBYNC4$$2HGFVOC$$aCreative Commons Attribution-NonCommercial CC BY-NC 4.0
000906936 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2021-01-28
000906936 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000906936 915__ $$0StatID:(DE-HGF)1150$$2StatID$$aDBCoverage$$bCurrent Contents - Physical, Chemical and Earth Sciences$$d2021-01-28
000906936 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2021-01-28
000906936 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2021-01-28
000906936 9201_ $$0I:(DE-Juel1)PGI-14-20210412$$kPGI-14$$lNeuromorphic Compute Nodes$$x0
000906936 980__ $$ajournal
000906936 980__ $$aVDB
000906936 980__ $$aUNRESTRICTED
000906936 980__ $$aI:(DE-Juel1)PGI-14-20210412
000906936 980__ $$aAPC
000906936 9801_ $$aAPC
000906936 9801_ $$aFullTexts