% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@INPROCEEDINGS{Yu:1022198,
author = {Yu, Zhenming and Menzel, Stephan and Strachan, John Paul
and Neftci, Emre},
title = {{I}ntegration of {P}hysics-{D}erived {M}emristor {M}odels
with {M}achine {L}earning {F}rameworks},
publisher = {IEEE},
reportid = {FZJ-2024-01319},
pages = {1142-1146},
year = {2022},
comment = {2022 56th Asilomar Conference on Signals, Systems, and
Computers : [Proceedings] - IEEE, 2022. - ISBN
978-1-6654-5906-8 - doi:10.1109/IEEECONF56349.2022.10052010},
booktitle = {2022 56th Asilomar Conference on
Signals, Systems, and Computers :
[Proceedings] - IEEE, 2022. - ISBN
978-1-6654-5906-8 -
doi:10.1109/IEEECONF56349.2022.10052010},
abstract = {Simulation 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.},
month = {Oct},
date = {2022-10-31},
organization = {2022 56th Asilomar Conference on
Signals, Systems, and Computers,
Pacific Grove (CA), 31 Oct 2022 - 2 Nov
2022},
cin = {PGI-15},
cid = {I:(DE-Juel1)PGI-15-20210701},
pnm = {5234 - Emerging NC Architectures (POF4-523) / BMBF
16ES1133K - Verbundprojekt: Neuro-inspirierte Technologien
der künstlichen Intelligenz für die Elektronik der Zukunft
- NEUROTEC -, Teilvorhaben: Forschungszentrum Jülich
(16ES1133K)},
pid = {G:(DE-HGF)POF4-5234 / G:(BMBF)16ES1133K},
typ = {PUB:(DE-HGF)8 / PUB:(DE-HGF)7},
UT = {WOS:000976687600210},
doi = {10.1109/IEEECONF56349.2022.10052010},
url = {https://juser.fz-juelich.de/record/1022198},
}