Conference paper
Reliability Challenges with Materials for Analog Computing
Eduard Cartier, Wanki Kim, et al.
IRPS 2019
We demonstrate a novel process for building a Resistive RAM (ReRAM) stack which reduces the forming voltage ( ) and increases the switching resistance, both characteristics that are important ingredients for the use of ReRAM in scalable analog compute for AI. Utilizing this process, we explore analog switching characteristics above 100k and demonstrate 4-bit programming at Rmax . Utilizing the same writing characteristics, CIFAR-10 inference simulation shows 90% accuracy, comparable to the full precision model accuracy.
Eduard Cartier, Wanki Kim, et al.
IRPS 2019
Youngseok Kim, Christopher J. Wood, et al.
Nature Physics
John H Zhang, Haigou Huang, et al.
MRS Advances
Kangguo Cheng, Ali Khakifirooz, et al.
CSTIC 2010