Date of Award
Electrical and Computer Engineering
Jinhui, Wang, Ph.D.
Dr. Samuel Russ, Dr. Jianing Han
Deep Neural Networks (DNN) are widely used in edge AI. But the complex perception and decision-making demand the overlarge computation and make the DNN architecture very sophisticated. Memristors have multilevel resistance property that enables faster in-memory DNN computation to remove the bottleneck caused by the von Neumann architecture and CMOS technology. However, the Stuck-At-Fault (SAF) defect of memristor generated from immature fabrication and heavy device utilization makes the memristor-based edge AI commercially unavailable. To mitigate this problem, an Adaptive Mapping Method (AMM) is proposed in this project. Based on the analysis for the VGG8 model with CIFAR10 dataset, the experiment results show that the AMM is efficient in restoring the inference accuracy up to 90% (the original accuracy without SAF) under SAFs from 0.1% to 50%, where Stuck-At-One (SAl): Stuck-At-Zero (SAO)= 5:1, 1:5, and 1:1. Additionally, the AMM has a significant immunity against the nonlinearity and conductance drift. The AMM improves the accuracy more than 50% in presence of high nonlinearity LTP = 4 and LTD = -4, and the standard conductance drift (10 years at 85 degree centigrade) nearly has no influence on the inference accuracy of the DNN in edge Al with the AMM.
Oli-Uz-Zaman, Md., "Stuck-At-Fault Immunity Enhancement of Memristor-Based Edge AI Systems" (2022). Theses and Dissertations. 115.
Available for download on Sunday, December 15, 2024
Artificial Intelligence and Robotics Commons, Computer and Systems Architecture Commons, Other Computer Engineering Commons