论文信息 - An in-the-loop training method for VLSI neural networks

An in-the-loop training method for VLSI neural networks

This paper deals with the in-the-loop training of an intelligent sensor that is based on the use of an artificial neural network with analog neurons, programmable digital weights and an integrated photosensitive array. In the training method, each of the neuron activation functions of the actual physical realization is measured and then modeled in terms of a small neural network. These small neural networks are embedded in a larger neural network that models the complete neural network. For the training of the complete neural network model, an algorithm that allows one to train the network when the analytic nature of both the nonlinear neuron activation function and its derivative are not known is presented. We also describe an approach to training the hardware implementation using digital weights of low resolution where the weight quantization effects are especially evident.

Majid Ahmadi | Graham A. Jullien | William C. Miller | J. Yang

[1] F. Girosi,et al. Networks for approximation and learning , 1990, Proc. IEEE.

[2] Marwan A. Jabri,et al. Weight perturbation: an optimal architecture and learning technique for analog VLSI feedforward and recurrent multilayer networks , 1992, IEEE Trans. Neural Networks.

[3] W. C. Miller,et al. Neuron modelling for in-the-loop training , 1998 .

[4] R. Palmer,et al. Introduction to the theory of neural computation , 1994, The advanced book program.

[5] Edward A. Rietman,et al. Back-propagation learning and nonidealities in analog neural network hardware , 1991, IEEE Trans. Neural Networks.

[6] Majid Ahmadi,et al. Model validation and determination for neural network activation function modeling , 1998, 1998 Midwest Symposium on Circuits and Systems (Cat. No. 98CB36268).

[7] Bernard Widrow,et al. 30 years of adaptive neural networks: perceptron, Madaline, and backpropagation , 1990, Proc. IEEE.

[8] Graham A. Jullien,et al. In-the-loop training algorithm for neural network implementation with digital weights , 1998, Other Conferences.

[9] L. Jones. Constructive approximations for neural networks by sigmoidal functions , 1990, Proc. IEEE.

[10] Majid Ahmadi,et al. A modular architecture for hybrid VLSI neural networks and its application in a smart photosensor , 1996, Proceedings of International Conference on Neural Networks (ICNN'96).