论文信息 - Precision requirements for single-layer feedforward neural networks

Precision requirements for single-layer feedforward neural networks

This paper presents a mathematical analysis of the effect of limited precision analog hardware for weight adaptation to be used in on-chip learning feedforward neural networks. Easy-to-read equations and simple worst-case estimations for the maximum tolerable imprecision are presented. As an application of the analysis, a worst-case estimation on the minimum size of the weight storage capacitors is presented.

Anne-Johan Annema | Klaas Hoen | Hans Wallinga

[1] G. Temes,et al. Analog MOS Integrated Circuits for Signal Processing , 1986 .

[2] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[3] Yiwen Wang. A modular analog CMOS LSI for feedforward neural networks with on-chip BEP learning , 1993, 1993 IEEE International Symposium on Circuits and Systems.

[4] Daniele D. Caviglia,et al. An experimental analog VLSI neural network with on-chip back-propagation learning , 1992, ESSCIRC '92: Eighteenth European Solid-State Circuits conference.

[5] J. J. Paulos,et al. On-chip learning in the analog domain with limited precision circuits , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[6] Daniele D. Caviglia,et al. Effects of weight discretization on the back propagation learning method: algorithm design and hardware realization , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[7] Anne-Johan Annema,et al. Learning behavior and temporary minima of two-layer neural networks , 1994, Neural Networks.

[8] Anne J. Annema. Analysis, Modelling and Implementation of Analog Integrated Neural Networks , 1994 .

[9] Torsten Lehmann. A Hardware Efficient Cascadable Chip Set For Ann'S With On-Chip Backpropagation , 1993, Int. J. Neural Syst..