001022198 001__ 1022198
001022198 005__ 20250129221932.0
001022198 0247_ $$2doi$$a10.1109/IEEECONF56349.2022.10052010
001022198 0247_ $$2datacite_doi$$a10.34734/FZJ-2024-01319
001022198 0247_ $$2WOS$$aWOS:000976687600210
001022198 037__ $$aFZJ-2024-01319
001022198 041__ $$aEnglish
001022198 1001_ $$0P:(DE-Juel1)190500$$aYu, Zhenming$$b0$$eCorresponding author$$ufzj
001022198 1112_ $$a2022 56th Asilomar Conference on Signals, Systems, and Computers$$cPacific Grove$$d2022-10-31 - 2022-11-02$$wCA
001022198 245__ $$aIntegration of Physics-Derived Memristor Models with Machine Learning Frameworks
001022198 260__ $$bIEEE$$c2022
001022198 29510 $$a2022 56th Asilomar Conference on Signals, Systems, and Computers : [Proceedings] - IEEE, 2022. - ISBN 978-1-6654-5906-8 - doi:10.1109/IEEECONF56349.2022.10052010
001022198 300__ $$a1142-1146
001022198 3367_ $$2ORCID$$aCONFERENCE_PAPER
001022198 3367_ $$033$$2EndNote$$aConference Paper
001022198 3367_ $$2BibTeX$$aINPROCEEDINGS
001022198 3367_ $$2DRIVER$$aconferenceObject
001022198 3367_ $$2DataCite$$aOutput Types/Conference Paper
001022198 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1738136748_6366
001022198 3367_ $$0PUB:(DE-HGF)7$$2PUB:(DE-HGF)$$aContribution to a book$$mcontb
001022198 520__ $$aSimulation frameworks such MemTorch, DNN+NeuroSim, and aihwkit are commonly used to facilitate the end-to-end co-design of memristive machine learning (ML) accelerators. These simulators can take device nonidealities into account and are integrated with modern ML frameworks. However, memristors in these simulators are modeled with either lookup tables or simple analytic models with basic nonlinearities. These simple models are unable to capture certain performance-critical aspects of device nonidealities. For example, they ignore the physical cause of switching, which induces errors in switching timings and thus incorrect estimations of conductance states. This work aims at bringing physical dynamics into consideration to model nonidealities while being compatible with GPU accelerators. We focus on Valence Change Memory (VCM) cells, where the switching nonlinearity and SET/RESET asymmetry relate tightly with the thermal resistance, ion mobility, Schottky barrier height, parasitic resistance, and other effects. The resulting dynamics require solving an ODE that captures changes in oxygen vacancies. We modified a physics-derived SPICE-level VCM model, integrated it with the aihwkit simulator and tested the performance with the MNIST dataset. Results show that noise that disrupts the SET/RESET matching affects network performance the most. This work serves as a tool for evaluating how physical dynamics in memristive devices affect neural network accuracy and can be used to guide the development of future integrated devices.
001022198 536__ $$0G:(DE-HGF)POF4-5234$$a5234 - Emerging NC Architectures (POF4-523)$$cPOF4-523$$fPOF IV$$x0
001022198 536__ $$0G:(BMBF)16ES1133K$$aBMBF 16ES1133K - Verbundprojekt: Neuro-inspirierte Technologien der künstlichen Intelligenz für die Elektronik der Zukunft - NEUROTEC -, Teilvorhaben: Forschungszentrum Jülich (16ES1133K)$$c16ES1133K$$x1
001022198 588__ $$aDataset connected to DataCite
001022198 7001_ $$0P:(DE-Juel1)158062$$aMenzel, Stephan$$b1$$ufzj
001022198 7001_ $$0P:(DE-Juel1)188145$$aStrachan, John Paul$$b2$$ufzj
001022198 7001_ $$0P:(DE-Juel1)188273$$aNeftci, Emre$$b3$$eCorresponding author$$ufzj
001022198 773__ $$a10.1109/IEEECONF56349.2022.10052010
001022198 8564_ $$uhttps://juser.fz-juelich.de/record/1022198/files/2403.06746.pdf$$yOpenAccess
001022198 8564_ $$uhttps://juser.fz-juelich.de/record/1022198/files/2403.06746.gif?subformat=icon$$xicon$$yOpenAccess
001022198 8564_ $$uhttps://juser.fz-juelich.de/record/1022198/files/2403.06746.jpg?subformat=icon-1440$$xicon-1440$$yOpenAccess
001022198 8564_ $$uhttps://juser.fz-juelich.de/record/1022198/files/2403.06746.jpg?subformat=icon-180$$xicon-180$$yOpenAccess
001022198 8564_ $$uhttps://juser.fz-juelich.de/record/1022198/files/2403.06746.jpg?subformat=icon-640$$xicon-640$$yOpenAccess
001022198 909CO $$ooai:juser.fz-juelich.de:1022198$$pVDB$$popen_access$$popenaire$$pdnbdelivery$$pdriver
001022198 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)190500$$aForschungszentrum Jülich$$b0$$kFZJ
001022198 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)158062$$aForschungszentrum Jülich$$b1$$kFZJ
001022198 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188145$$aForschungszentrum Jülich$$b2$$kFZJ
001022198 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188273$$aForschungszentrum Jülich$$b3$$kFZJ
001022198 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$$x0
001022198 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001022198 920__ $$lyes
001022198 9201_ $$0I:(DE-Juel1)PGI-15-20210701$$kPGI-15$$lNeuromorphic Software Eco System$$x0
001022198 980__ $$acontrib
001022198 980__ $$aVDB
001022198 980__ $$acontb
001022198 980__ $$aI:(DE-Juel1)PGI-15-20210701
001022198 980__ $$aUNRESTRICTED
001022198 9801_ $$aFullTexts