CONFERENCE PROCEEDINGS
Advanced Search
Home > Proceedings > Volume PC13119 > Article
Presentation
4 October 2024 Method for training robust neural network solutions for MTJ-based hardware with defects
William Borders, Advait Madhavan, Matthew Daniels, Vasileia Georgiou, Martin Lueker-Boden, Tiffany Santos, Patrick Braganca, Mark Stiles, Jabez McClelland, Brian Hoskins
Author Affiliations +
Proceedings Volume PC13119, Spintronics XVII; PC131190S (2024) https://doi.org/10.1117/12.3027145
Event: Nanoscience + Engineering, 2024, San Diego, California, United States
Abstract
Neural networks are increasing in scale and sophistication, catalyzing the need for efficient hardware. An inevitability when transferring neural networks to hardware is that non-idealities impact performance. Hardware-aware training, where non-idealities are accounted for during training is one way to recover performance, but at the cost of generality. In this work, we demonstrate a binary neural network consisting of an array of 20,000 magnetic tunnel junctions (MTJ) integrated on complementary metal-oxide-semiconductor (CMOS) chips. With 36 dies, we show that even a few defects can degrade the performance of neural networks. We demonstrate hardware-aware training and show that performance recovers close to ideal networks. We then introduce a robust method – statistics-aware training – that compensates for defects regardless of their specific configuration. When evaluated on the MNIST dataset, statistics-aware solutions differ from software-baselines by only 2 %. We quantify the sensitivity of networks trained with statistics-aware and conventional methods and demonstrate that the statistics-aware solution shows less sensitivity to defects when sampling the network loss function.
Conference Presentation
© (2024) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Citation Download Citation
William Borders, Advait Madhavan, Matthew Daniels, Vasileia Georgiou, Martin Lueker-Boden, Tiffany Santos, Patrick Braganca, Mark Stiles, Jabez McClelland, and Brian Hoskins "Method for training robust neural network solutions for MTJ-based hardware with defects", Proc. SPIE PC13119, Spintronics XVII, PC131190S (4 October 2024); https://doi.org/10.1117/12.3027145
ACCESS THE FULL ARTICLE
ORGANIZATIONAL
Sign in with credentials provided by your organization.
INSTITUTIONAL
Select your institution to access the SPIE Digital Library.
SELECT YOUR INSTITUTION
PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
PERSONAL SIGN IN
No SPIE Account? Create one
PURCHASE THIS CONTENT
SUBSCRIBE TO DIGITAL LIBRARY
50 downloads per 1-year subscription
Members: $195
Non-members: $335 ADD TO CART
25 downloads per 1 - year subscription
Members: $145
Non-members: $250 ADD TO CART
PURCHASE SINGLE ARTICLE
Includes PDF, HTML & Video, when available
Members: $17.00
Non-members: $21.00 ADD TO CART
PROCEEDINGS
 PRESENTATION
WATCH
PRESENTATION SAVE TO MY LIBRARY
GET CITATION
Advertisement
Advertisement
KEYWORDS
Education and training
Artificial neural networks
Computer hardware
Binary data
Industrial applications
Magnetic tunnel junctions
RELATED CONTENT
Subscribe to Digital Library
Receive Erratum Email Alert
Back to Top