Highly Reliable Synaptic Cell Array Based on Organic–Inorganic Hybrid Bilayer Stack toward Precise Offline Learning

Advanced Intelligent Systems

Sung Gap Im¹  Changhyeon Lee¹  Hamin Park²  Sung-Yool Choi³  Sang Yoon Yang³  Jun-Hwe Cha³  Jungyeop Oh³  Byung Chul Jang⁴ 

[1] Department of Chemical and Biomolecular Engineering Korea Advanced Institute of Science and Technology (KAIST) 291 Daehak-ro Yuseong-gu Daejeon 34141 Republic of Korea;Department of Electronic Engineering Kwangwoon University 20 Gwangun-ro Nowon-gu Seoul 01897 Republic of Korea;School of Electrical Engineering Graphene/2D Materials Research Center Center for Advanced Materials Discovery towards 3D Displays Korea Advanced Institute of Science and Technology (KAIST) 291 Daehak-ro Yuseong-gu Daejeon 34141 Republic of Korea;School of Electronics Engineering Kyungpook National University 80 Daehakro Bukgu Daegu 41566 Republic of Korea;

关键词: 5-bit multilevel retention;    conductive-bridging random-access memory (CBRAM);    fine-tuning;    neuromorphic computing;    offline learning;   

DOI  :  10.1002/aisy.202200018

来源: DOAJ

【 摘 要 】

As the use of artificial intelligence (AI) soars, the development of novel neuromorphic computing is demanding because of the disadvantages of the von Neumann architecture. Furthermore, extensive research on electrochemical metallization (ECM) memristors as synaptic cells have been carried out toward a linear conductance update for online learning applications. In most cases, however, a conductance distribution change over time has not been studied as a major issue, giving less consideration to inference‐only computing accelerators based on offline learning. Herein, organic–inorganic bilayer stacking for synaptic unit cells using poly(1,3,5‐trivinyl‐1,3,5‐trimethyl cyclotrisiloxane) (pV3D3) and Al2O3 thin films is suggested, showing highly enhanced reliability for offline learning. The bilayer structure achieves better reliability and control of the analog resistive switching and synaptic functions, respectively, through the guided formation of conductive filaments via tip‐enhanced electric fields. In addition, 5‐bit multilevel states achieve long‐term stability (>104 s) following an in‐depth study on conductance‐level stability. Finally, a device‐to‐system‐level simulation is performed by building a convolutional neural network (CNN) based on the hybrid devices. This highlighted the significance of multilevel states in fully connected layers. It is believed that the study provides a practical approach to using ECM‐based memristors for inference‐only neural network accelerators.

【 授权许可】

Unknown