Room‐Temperature‐Processable Highly Reliable Resistive Switching Memory with Reconfigurability for Neuromorphic Computing and Ultrasonic Tissue Classification. (17th January 2023)

• Record Type: Journal Article
• Authors: Kim, Dohyung; Bang, Hyeonsu; Baac, Hyoung Won; Lee, Jongmin; Truong, Phuoc Loc; Jeong, Bum Ho; Appadurai, Tamilselvan; Park, Kyu Kwan; Heo, Donghyeok; Nam, Vu Binh; Yoo, Hocheon; Kim, Kyeounghak; Lee, Daeho; Ko, Jong Hwan; Park, Hui Joon
• Is Part Of: Advanced functional materials. Volume 33:Number 14(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-01-17
• Subjects: 3D ion transport channels -- flexible -- hardware learning rules -- lasers -- neuromorphic -- reconfigurable -- resistive switching memory -- synapses; Materials -- Periodicals; Chemical vapor deposition -- Periodicals; 620.11
• DOI: 10.1002/adfm.202213064
• Languages: English
• ISSNs: 1616-301X

Abstract

Abstract: Reversible metal‐filamentary mechanism has been widely investigated to design an analog resistive switching memory (RSM) for neuromorphic hardware‐implementation. However, uncontrollable filament‐formation, inducing its reliability issues, has been a fundamental challenge. Here, an analog RSM with 3D ion transport channels that can provide unprecedentedly high reliability and robustness is demonstrated. This architecture is realized by a laser‐assisted photo‐thermochemical process, compatible with the back‐end‐of‐line process and even applicable to a flexible format. These superior characteristics also lead to the proposal of a practical adaptive learning rule for hardware neural networks that can significantly simplify the voltage pulse application methodology even with high computing accuracy. A neural network, which can perform the biological tissue classification task using the ultrasound signals, is designed, and the simulation results confirm that this practical adaptive learning rule is efficient enough to classify these weak and complicated signals with high accuracy (97%). Furthermore, the proposed RSM can work as a diffusive‐memristor at the opposite voltage polarity, exhibiting extremely stable threshold switching characteristics. In this mode, several crucial operations in biological nervous systems, such as Ca 2+ dynamics and nonlinear integrate‐and‐fire functions of neurons, are successfully emulated. This reconfigurability is also exceedingly beneficial for decreasing the complexity of systems—requiring both drift‐ and diffusive‐memristors. Abstract : Analog resistive switching memory (RSM) with unprecedentedly high reliability and robustness is demonstrated by a laser‐assisted photo‐thermochemical process. This architecture is compatible with back‐end‐of‐line process and flexible format. Based on its superior characteristics, a practical adaptive learning rule is designed and applied to ultrasonic tissue classification task with high computing accuracy. The proposed RSM also has reconfigurability working as a diffusive‐memristor.