Recognition Accuracy Enhancement using Interface Control with Weight Variation-Lowering in Analog Computation-in-Memory

As AI technology develops, it is necessary to verify the technical feasibility of Memory-Centric convergence technology. Previously investigated resistive synaptic devices (RSDs) can successfully mimic the function of biological synapses. However, the effect of the system recognition rate reflecting...

Full description

Saved in:
Bibliographic Details
Main Authors: Park, Sangsu, Lee, Gyonhui, Kwon, Youngjae, Suh, Dong Ik, Lee, Hanwool, Je, Sangeun, Kim, Dabin, Lee, Dohan, Ryu, Seungwook, Kim, Seungbum, Kim, Euiseok, Lee, Sunghoon, Park, Kyoung, Lee, Seho, Na, Myung-Hee, Cha, Seonyong
Format: Conference Proceeding
Language:English
Subjects:
Analog Computation-in-Memory (ACiM)
Artificial intelligence
Biological system modeling
Computational modeling
Conductance Variation
Conferences
interface control
multiplication-and-accumulation (MAC)
Optimization methods
Performance evaluation
Resistive Synaptic Device (RSD)
Simulation
Weight variation-lowering
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:As AI technology develops, it is necessary to verify the technical feasibility of Memory-Centric convergence technology. Previously investigated resistive synaptic devices (RSDs) can successfully mimic the function of biological synapses. However, the effect of the system recognition rate reflecting the variation of 16 weight states has not been studied yet. In this article, we perform simulations of various weight variation sets through real resistive synaptic device (RSD) engineering in Analog Computation-in-Memory (ACiM) system. These simulation results can provide guidelines for the continued design and optimization of a resistive synaptic device for realizing ACiM.
ISSN:2573-7503
DOI:10.1109/IMW52921.2022.9779296