001     1025209
005     20250203103227.0
024 7 _ |a 10.29363/nanoge.neuronics.2024.027
|2 doi
037 _ _ |a FZJ-2024-02777
100 1 _ |a Neftci, Emre
|0 P:(DE-Juel1)188273
|b 0
|u fzj
111 2 _ |a Neuronics Conference
|c València
|d 2024-02-21 - 2024-02-23
|w Spain
245 _ _ |a Pre-training and Meta-learning for Memristor Crossbar Arrays
260 _ _ |c 2023
336 7 _ |a Conference Paper
|0 33
|2 EndNote
336 7 _ |a Other
|2 DataCite
336 7 _ |a INPROCEEDINGS
|2 BibTeX
336 7 _ |a conferenceObject
|2 DRIVER
336 7 _ |a LECTURE_SPEECH
|2 ORCID
336 7 _ |a Conference Presentation
|b conf
|m conf
|0 PUB:(DE-HGF)6
|s 1714556807_3947
|2 PUB:(DE-HGF)
|x Invited
520 _ _ |a Memristive crossbar arrays show promise as non-von Neumann computing technologies, bringing sophisticated neural network processing to the edge and facilitating real-world online learning. However, their deployment for real-world learning problems faces challenges such as non-linearities in conductance updates, variations during operation, fabrication mismatch, conductance drift, and the realities of gradient descent training.This talk will present methods to pre-train neural networks to be largely insensitive to these non-idealities during learning tasks. These methods rely on a phenomenological model of the device, obtainable experimentally, and bi-level optimization. We showcase this effect through meta-learning and a differentiable model of conductance updates on few-shot learning tasks. Since pre-training is a necessary procedure for any online learning scenario at the edge, our results may pave the way for real-world applications of memristive devices without significant adaptation overhead.Furthermore, by considering the programming of memristive devices as a learning problem in its own right, we demonstrate that the developed methods can accelerate existing write-verify techniques.
536 _ _ |a 5234 - Emerging NC Architectures (POF4-523)
|0 G:(DE-HGF)POF4-5234
|c POF4-523
|f POF IV
|x 0
588 _ _ |a Dataset connected to CrossRef Conference
773 _ _ |a 10.29363/nanoge.neuronics.2024.027
856 4 _ |u https://www.nanoge.org/proceedings/Neuronics/656741f369ece060987dcf51
909 C O |o oai:juser.fz-juelich.de:1025209
|p VDB
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 0
|6 P:(DE-Juel1)188273
913 1 _ |a DE-HGF
|b Key Technologies
|l Natural, Artificial and Cognitive Information Processing
|1 G:(DE-HGF)POF4-520
|0 G:(DE-HGF)POF4-523
|3 G:(DE-HGF)POF4
|2 G:(DE-HGF)POF4-500
|4 G:(DE-HGF)POF
|v Neuromorphic Computing and Network Dynamics
|9 G:(DE-HGF)POF4-5234
|x 0
914 1 _ |y 2024
920 1 _ |0 I:(DE-Juel1)PGI-15-20210701
|k PGI-15
|l Neuromorphic Software Eco System
|x 0
980 _ _ |a conf
980 _ _ |a VDB
980 _ _ |a I:(DE-Juel1)PGI-15-20210701
980 _ _ |a UNRESTRICTED


LibraryCollectionCLSMajorCLSMinorLanguageAuthor
Marc 21