No abstract

No abstract

No abstract

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.

Analog in-memory computing synaptic devices have been widely studied for efficient implementation of deep learning. As a candidate for a synaptic device, Si-CMOS and capacitor-based synaptic devices have been proposed. However, due to Si-CMOS leakage currents, it is difficult to achieve sufficient retention time. In our research, we verified IGZO TFT with low leakage current and capacitor-based synapses can show linear and symmetric weight update characteristics as well as excellent device variation characteristics. We also verified that IGZO TFT has a leakage current per channel width of 1μm of ~10-17A, which is much lower than the Si-CMOS, resulting in higher accuracy in deep neural network training.