Resonance for Analog Recurrent Neural Network

被引：8

作者：

Qu, Yurui ^{[1
]}

Zhou, Ming ^{[1
]}

Khoram, Erfan ^{[1
]}

Yu, Nanfang ^{[1
]}

Yu, Zongfu ^{[1
]}

机构：

[1] Univ Wisconsin, Dept Elect & Comp Engn, Madison, WI 53706 USA

来源：

ACS PHOTONICS | 2022年 / 9卷 / 05期

关键词：

resonance; temporal memory; recurrent neural networks; vowel classification; LSTM; INVERSE DESIGN;

D O I：

10.1021/acsphotonics.1c02016

中图分类号：

TB3 [工程材料学];

学科分类号：

0805 ; 080502 ;

摘要：

There is a strong interest in using physical waves for artificial neuralcomputing because of their unique advantages in fast speed and intrinsicparallelism. Resonance, as a ubiquitous feature across many wave systems, is anatural candidate for analog computing in temporal signals. We demonstrate thatresonance can be used to construct stable and scalable recurrent neural networks.By including resonators with different lifetimes, the computing system developsboth short-term and long-term memories simultaneously

引用

页码：1647 / 1654

页数：8

共 50 条

[31] A recurrent neural network that learns to count
Rodriguez, P
Wiles, J
Elman, JL
CONNECTION SCIENCE, 1999, 11 (01) : 5 - 40
[32] Stochastic graph recurrent neural network
Yan, Tijin
Zhang, Hongwei
Li, Zirui
Xia, Yuanqing
NEUROCOMPUTING, 2022, 500 : 1003 - 1015
[33] Kp forecasting with a recurrent neural network
Sexton, Ernest Scott
Nykyri, Katariina
Ma, Xuanye
JOURNAL OF SPACE WEATHER AND SPACE CLIMATE, 2019, 9
[34] Edge detection with a recurrent neural network
Vrabel, MJ
APPLICATIONS AND SCIENCE OF ARTIFICIAL NEURAL NETWORKS II, 1996, 2760 : 365 - 371
[35] MODEL OF A NEURAL NETWORK WITH RECURRENT INHIBITION
WIGSTROM, H
KYBERNETIK, 1974, 16 (02): : 103 - 112
[36] Recurrent neural network wave functions
Hibat-Allah, Mohamed
Ganahl, Martin
Hayward, Lauren E.
Melko, Roger G.
Carrasquilla, Juan
PHYSICAL REVIEW RESEARCH, 2020, 2 (02):
[37] Time Adaptive Recurrent Neural Network
Kag, Anil
Saligrama, Venkatesh
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 15144 - 15153
[38] Recurrent Neural Network with Dynamic Memory
Bai, Jiaqi
Dong, Tao
Liao, Xiaofeng
Mu, Nankun
ADVANCES IN NEURAL NETWORKS - ISNN 2018, 2018, 10878 : 339 - 345
[39] A RECURRENT NEURAL NETWORK - LIMITATIONS AND TRAINING
LEVIN, E
PROCEEDINGS OF THE 22ND CONFERENCE ON INFORMATION SCIENCES AND SYSTEMS, VOLS 1 & 2, 1988, : 296 - 301
[40] A Dynamically Stabilized Recurrent Neural Network
Samer Saab
Yiwei Fu
Asok Ray
Michael Hauser
Neural Processing Letters, 2022, 54 : 1195 - 1209

← 1 2 3 4 5 →