中国科学:物理学 力学 天文学(英文版)2024，Vol.67Issue(8) ：23-42.DOI:10.1007/s11433-024-2403-x

Physical neural networks with self-learning capabilities

Weichao Yu Hangwen Guo Jiang Xiao Jian Shen
中国科学:物理学 力学 天文学(英文版)2024，Vol.67Issue(8) ：23-42.DOI:10.1007/s11433-024-2403-x
✕
  • 维普
  • 万方数据

Physical neural networks with self-learning capabilities

Weichao Yu 1Hangwen Guo 2Jiang Xiao 3Jian Shen3
扫码查看

作者信息

  • 1. State Key Laboratory of Surface Physics and Institute for Nanoelectronic Devices and Quantum Computing,Fudan University,Shanghai 200433,China;Zhangjiang Fudan International Innovation Center,Fudan University,Shanghai 201210,China
  • 2. State Key Laboratory of Surface Physics and Institute for Nanoelectronic Devices and Quantum Computing,Fudan University,Shanghai 200433,China;Zhangjiang Fudan International Innovation Center,Fudan University,Shanghai 201210,China;Shanghai Research Center for Quantum Sciences,Shanghai 201315,China
  • 3. State Key Laboratory of Surface Physics and Institute for Nanoelectronic Devices and Quantum Computing,Fudan University,Shanghai 200433,China;Zhangjiang Fudan International Innovation Center Fudan University,Shanghai 201210,China;Department of Physics,Fudan University,Shanghai 200433,China;Shanghai Research Center for Quantum Sciences,Shanghai 201315,China;Collaborative Innovation Center of Advanced Microstructures,Nanjing 210093,China
  • 折叠

Abstract

Physical neural networks are artificial neural networks that mimic synapses and neurons using physical systems or materials.These networks harness the distinctive characteristics of physical systems to carry out computations effectively,potentially sur-passing the constraints of conventional digital neural networks.A recent advancement known as"physical self-learning"aims to achieve learning through intrinsic physical processes rather than relying on external computations.This article offers a compre-hensive review of the progress made in implementing physical self-learning across various physical systems.Prevailing learning strategies that contribute to the realization of physical self-learning are discussed.Despite challenges in understanding the funda-mental mechanism of learning,this work highlights the progress towards constructing intelligent hardware from the ground up,incorporating embedded self-organizing and self-adaptive dynamics in physical systems.

Key words

self-learning/physical neural networks/neuromorphic computing/physical learning

引用本文复制引用

基金项目

National Key Research and Development Program of China(2022YFA1403300)

National Key Research and Development Program of China(2020YFA0309100)

National Natural Science Foundation of China(12204107)

National Natural Science Foundation of China(12074073)

Shanghai Municipal Science and Technology Major Project(2019SHZDZX01)

Shanghai Pujiang Program(21PJ1401500)

Shanghai Science and Technology Committee(21JC1406200)

Shanghai Science and Technology Committee(20JC1415900)

出版年

2024
中国科学:物理学 力学 天文学(英文版)
中国科学院

中国科学:物理学 力学 天文学(英文版)

CSTPCD
影响因子：0.91
ISSN：1674-7348
段落导航相关论文