Analog CMOS neural networks based on Gilbert multipliers with in-circuit learning
暂无分享,去创建一个
This paper examines analog CMOS circuit implementations of several common neural network algorithms. All circuits described perform in-circuit learning, using Gilbert multipliers as a primary circuit component. These include 3 /spl mu/m and 1.2 /spl mu/m designs for contrastive Hebbian learning, and Becker-Hinton networks (a variation of delta-rule learning). In addition, unsupervised learning circuits for competitive learning are presented.<<ETX>>
[1] Helen Suzanna Becker,et al. An information-theoretic unsupervised learning algorithm for neural networks , 1993 .
[2] Barrie Gilbert. A high-performance monolithic multiplier using active feedback , 1974 .
[3] Geoffrey E. Hinton,et al. Self-organizing neural network that discovers surfaces in random-dot stereograms , 1992, Nature.
[4] Karsten P. Ulland,et al. Vii. References , 2022 .