000972102 001__ 972102
000972102 005__ 20230929112510.0
000972102 0247_ $$2doi$$a10.3389/femat.2023.1061269
000972102 0247_ $$2Handle$$a2128/33847
000972102 037__ $$aFZJ-2023-01070
000972102 082__ $$a540
000972102 1001_ $$0P:(DE-Juel1)188159$$aBengel, Christopher$$b0$$eCorresponding author
000972102 245__ $$aTailor-made synaptic dynamics based on memristive devices
000972102 260__ $$aLausanne$$bFrontiers Media$$c2023
000972102 3367_ $$2DRIVER$$aarticle
000972102 3367_ $$2DataCite$$aOutput Types/Journal article
000972102 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1675236225_7868
000972102 3367_ $$2BibTeX$$aARTICLE
000972102 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000972102 3367_ $$00$$2EndNote$$aJournal Article
000972102 520__ $$aThe proliferation of machine learning algorithms in everyday applications such as image recognition or language translation has increased the pressure to adapt underlying computing architectures towards these algorithms. Application specific integrated circuits (ASICs) such as the Tensor Processing Units by Google, Hanguang by Alibaba or Inferentia by Amazon Web Services were designed specifically for machine learning algorithms and have been able to outperform CPU based solutions by great margins during training and inference. As newer generations of chips allow handling of and computation on more and more data, the size of neural networks has dramatically increased, while the challenges they are trying to solve have become more complex. Neuromorphic computing tries to take inspiration from biological information processing systems, aiming to further improve the efficiency with which these networks can be trained or the inference can be performed. Enhancing neuromorphic computing architectures with memristive devices as non-volatile storage elements could potentially allow for even higher energy efficiencies. Their ability to mimic synaptic plasticity dynamics brings neuromorphic architectures closer to the biological role models. So far, memristive devices are mainly investigated for the emulation of the weights of neural networks during training and inference as their non-volatility would enable both processes in the same location without data transfer. In this paper, we explore realisations of different synapses build from memristive ReRAM devices, based on the Valence Change Mechanism. These synapses are the 1R synapse, the NR synapse and the 1T1R synapse. For the 1R synapse, we propose three dynamical regimes and explore their performance through different synapse criteria. For the NR synapse, we discuss how the same dynamical regimes can be addressed in a more reliable way. We also show experimental results measured on ZrOx devices to support our simulation based claims. For the 1T1R synapse, we explore the trade offs between the connection direction of the ReRAM device and the transistor. For all three synapse concepts we discuss the impact of device-to-device and cycle-to-cycle variability. Additionally, the impact of the stimulation mode on the observed behavior is discussed.
000972102 536__ $$0G:(DE-HGF)POF4-5233$$a5233 - Memristive Materials and Devices (POF4-523)$$cPOF4-523$$fPOF IV$$x0
000972102 536__ $$0G:(DE-82)BMBF-16ME0399$$aBMBF-16ME0399 - Verbundprojekt: Neuro-inspirierte Technologien der künstlichen Intelligenz für die Elektronik der Zukunft - NEUROTEC II - (BMBF-16ME0399)$$cBMBF-16ME0399$$x1
000972102 536__ $$0G:(DE-82)BMBF-16ME0398K$$aBMBF-16ME0398K - Verbundprojekt: Neuro-inspirierte Technologien der künstlichen Intelligenz für die Elektronik der Zukunft - NEUROTEC II - (BMBF-16ME0398K)$$cBMBF-16ME0398K$$x2
000972102 536__ $$0G:(DE-HGF)SO-092$$aACA - Advanced Computing Architectures (SO-092)$$cSO-092$$x3
000972102 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
000972102 7001_ $$0P:(DE-HGF)0$$aZhang, Kaihua$$b1
000972102 7001_ $$0P:(DE-HGF)0$$aMohr, Johannes$$b2
000972102 7001_ $$0P:(DE-Juel1)177689$$aZiegler, Tobias$$b3
000972102 7001_ $$0P:(DE-Juel1)187229$$aWiefels, Stefan$$b4
000972102 7001_ $$0P:(DE-Juel1)131022$$aWaser, R.$$b5
000972102 7001_ $$0P:(DE-HGF)0$$aWouters, Dirk$$b6
000972102 7001_ $$0P:(DE-Juel1)158062$$aMenzel, Stephan$$b7
000972102 773__ $$0PERI:(DE-600)3106175-8$$a10.3389/femat.2023.1061269$$gVol. 3, p. 1061269$$p1061269$$tFrontiers in electronic materials$$v3$$x2673-9895$$y2023
000972102 8564_ $$uhttps://juser.fz-juelich.de/record/972102/files/femat-03-1061269.pdf$$yOpenAccess
000972102 8767_ $$d2023-04-04$$eAPC$$jDeposit$$z1768 USD
000972102 909CO $$ooai:juser.fz-juelich.de:972102$$pdnbdelivery$$popenCost$$pVDB$$pdriver$$pOpenAPC$$popen_access$$popenaire
000972102 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188159$$aForschungszentrum Jülich$$b0$$kFZJ
000972102 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)177689$$aForschungszentrum Jülich$$b3$$kFZJ
000972102 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)187229$$aForschungszentrum Jülich$$b4$$kFZJ
000972102 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)131022$$aForschungszentrum Jülich$$b5$$kFZJ
000972102 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)158062$$aForschungszentrum Jülich$$b7$$kFZJ
000972102 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)POF4-5233$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vNeuromorphic Computing and Network Dynamics$$x0
000972102 9141_ $$y2023
000972102 915pc $$0PC:(DE-HGF)0000$$2APC$$aAPC keys set
000972102 915pc $$0PC:(DE-HGF)0001$$2APC$$aLocal Funding
000972102 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000972102 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
000972102 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2022-11-10T09:54:54Z
000972102 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2022-11-10T09:54:54Z
000972102 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Anonymous peer review$$d2022-11-10T09:54:54Z
000972102 9201_ $$0I:(DE-Juel1)PGI-7-20110106$$kPGI-7$$lElektronische Materialien$$x0
000972102 9201_ $$0I:(DE-82)080009_20140620$$kJARA-FIT$$lJARA-FIT$$x1
000972102 9201_ $$0I:(DE-Juel1)PGI-10-20170113$$kPGI-10$$lJARA Institut Green IT$$x2
000972102 9801_ $$aFullTexts
000972102 980__ $$ajournal
000972102 980__ $$aVDB
000972102 980__ $$aUNRESTRICTED
000972102 980__ $$aI:(DE-Juel1)PGI-7-20110106
000972102 980__ $$aI:(DE-82)080009_20140620
000972102 980__ $$aI:(DE-Juel1)PGI-10-20170113
000972102 980__ $$aAPC