Fast and robust analog in-memory deep neural network training

Malte J. Rasch (), Fabio Carta, Omobayode Fagbohungbe and Tayfun Gokmen ()
Additional contact information
Malte J. Rasch: TJ Watson Research Center
Fabio Carta: TJ Watson Research Center
Omobayode Fagbohungbe: TJ Watson Research Center
Tayfun Gokmen: TJ Watson Research Center

Nature Communications, 2024, vol. 15, issue 1, 1-15

Abstract: Abstract Analog in-memory computing is a promising future technology for efficiently accelerating deep learning networks. While using in-memory computing to accelerate the inference phase has been studied extensively, accelerating the training phase has received less attention, despite its arguably much larger compute demand to accelerate. While some analog in-memory training algorithms have been suggested, they either invoke significant amount of auxiliary digital compute—accumulating the gradient in digital floating point precision, limiting the potential speed-up—or suffer from the need for near perfectly programming reference conductance values to establish an algorithmic zero point. Here, we propose two improved algorithms for in-memory training, that retain the same fast runtime complexity while resolving the requirement of a precise zero point. We further investigate the limits of the algorithms in terms of conductance noise, symmetry, retention, and endurance which narrow down possible device material choices adequate for fast and robust in-memory deep neural network training.

Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-024-51221-z Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:15:y:2024:i:1:d:10.1038_s41467-024-51221-z

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-024-51221-z

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().