Hardware Implementation of the Multifrequency Oscillation Learning Method for Analog Neural Networks
スポンサーリンク
概要
- 論文の詳細を見る
High-speed learning of neural networks using the multifrequency oscillation method is demonstrated for first time. During the learning of an analog neural network integrated circuit implementing the `exclusive-OR' logic, weight and threshold values converge to steady states within 2 ms for a learning speed of 2 mega-patterns per second.
- 社団法人電子情報通信学会の論文
- 1993-06-25
著者
-
Koga M
Ntt Corp. Yokosuka‐shi Jpn
-
Koga Masafumi
Ntt Corporation
-
Koga M
Hitachi Ltd. Kokubunji‐shi Jpn
-
Koga Masafumi
Ntt Transmission Systems Laboratories
-
Matsumoto T
Waseda Univ. Tokyo Jpn
-
Miyao H
Ntt Optical Network Systems Lab. Yokosuka‐shi Jpn
-
Miyao Hiroshi
NTT Transmission Systems Laboratories
-
Matsumoto Takao
NTT Transmission Systems Laboratories
関連論文
- Flexible OADM Architecture and Its Impact on WDM Ring Evolution for Robust and Large-Scale Optical Transport Networks (Joint Special Issue on Recent Progress in Optoelectronics and Communications)
- Flexible OADM Architecture and Its Impact on WDM Ring Evolution for Robust and Large-Scale Optical Transport Networks (Joint Special Issue on Recent Progress in Optoelectronics and Communications)
- Design and Performance of Delivery and Coupling Switch Board for Large Scale Optical Path Cross-Connect System
- Hardware Implementation of the Multifrequency Oscillation Learning Method for Analog Neural Networks
- Fast Restoration on Network Control Plane Established through Photonic MPLS Routers(The IEICE Transactions on Communications, Vol.E86-B, No.5)
- Fast Restoration on Network Control Plane Established through Photonic MPLS Routers ( Recent Progress in Optoelectronics and Communications)
- Very-High-Speed Analog Neural Network LSI Using Super Self-Aligned Si Bipolar Process Technology