Journal Article

Device‐Algorithm Co‐Optimization for an On‐Chip Trainable Capacitor‐Based Synaptic Device with IGZO TFT and Retention‐Centric Tiki‐Taka Algorithm

Lee, Kwang‐Hee, Kim, Sangbum, Kang, Minil, Yang, Jee‐Eun, Oh, Saeroonter, Seo, Hyeong Jun, Han, Narae, Won, Jongun, Lee, Hyung‐Min, Kang, Minseung, Kang, Jaehyeon, Hong, Sangjun, Cho, Ung, Park, Yeaji, Roh, Youngchae, Joe, Changhoon, Um, Minseong, Jung, Moonil, Kim, Sangwook

2023

Overview

Analog in‐memory computing synaptic devices are widely studied for efficient implementation of deep learning. However, synaptic devices based on resistive memory have difficulties implementing on‐chip training due to the lack of means to control the amount of resistance change and large device variations. To overcome these shortcomings, silicon complementary metal‐oxide semiconductor (Si‐CMOS) and capacitor‐based charge storage synapses are proposed, but it is difficult to obtain sufficient retention time due to Si‐CMOS leakage currents, resulting in a deterioration of training accuracy. Here, a novel 6T1C synaptic device using only n‐type indium gaIlium zinc oxide thin film transistor (IGZO TFT) with low leakage current and a capacitor is proposed, allowing not only linear and symmetric weight update but also sufficient retention time and parallel on‐chip training operations. In addition, an efficient and realistic training algorithm to compensate for any remaining device non‐idealities such as drifting references and long‐term retention loss is proposed, demonstrating the importance of device‐algorithm co‐optimization. A novel 6T1C synaptic device based on indium gallium zinc oxide thin film transistor (IGZO TFT) and capacitor and a novel optimized training algorithm, retention‐centric Tiki‐Taka algorithm, is proposed. Through a new training scheme by co‐optimizing the device and algorithm, modified national institute of standards and technology (MNIST) on‐chip training accuracy of over ≈97% even in wide retention requirements is obtained.

• Publisher: John Wiley & Sons, Inc,John Wiley and Sons Inc,Wiley
• Subject: Algorithms / device‐algorithm co‐optimization / indium gallium zinc oxide thin film transistor (IGZO TFT) / in‐memory computing / Neural networks / neuromorphic / Random access memory / Retention / Semiconductors / Silicon wafers / Symmetry / tiki‐taka algorithm / Transistors