Neural Network Training With Asymmetric Crosspoint Elements

Neural Network Training With Asymmetric Crosspoint Elements

Analog crossbar arrays comprising programmable non-volatile resistors are under intense investigation for acceleration of deep neural network training. However, the ubiquitous asymmetric conductance modulation of practical resistive devices critically degrades the classification performance of netwo...

Full description

Bibliographic Details
Main Authors: Murat Onen, Tayfun Gokmen, Teodor K. Todorov, Tomasz Nowicki, Jesús A. del Alamo, John Rozen, Wilfried Haensch, Seyoung Kim
Format: Article
Language:English
Published: Frontiers Media S.A. 2022-05-01
Series:Frontiers in Artificial Intelligence
Subjects:
analog computing
DNN training
hardware accelerator architecture
neuromorphic accelerator
learning algorithm
Online Access:https://www.frontiersin.org/articles/10.3389/frai.2022.891624/full

Internet

https://www.frontiersin.org/articles/10.3389/frai.2022.891624/full

Similar Items