论文信息 - Training a perceptron by a bit sequence: storage capacity

Training a perceptron by a bit sequence: storage capacity

A perceptron is trained by a random bit sequence. In comparison with the corresponding classification problem, the storage capacity decreases to due to correlations between input and output bits. The numerical results are supported by a signal-to-noise analysis of Hebbian weights.

I. Kanter | W. Kinzel | M. Schroeder

[1] Wolfgang Kinzel,et al. Models of neural networks , 1987 .

[2] B. Harris. PROBABILITY DISTRIBUTIONS RELATED TO RANDOM MAPPINGS , 1960 .

[3] Andreas S. Weigend,et al. Time Series Prediction: Forecasting the Future and Understanding the Past , 1994 .

[4] M. Opper,et al. Storage of correlated patterns in a perceptron , 1995 .

[5] Bernard Derrida,et al. The random map model: a disordered model with deterministic dynamics , 1987 .

[6] Kanter,et al. Analytical study of time series generation by feed-forward networks. , 1995, Physical review letters.

[7] Eytan Domany,et al. Models of Neural Networks I , 1991 .

[8] Kanter,et al. Generation and prediction of time series by a neural network. , 1995, Physical review letters.

[9] Anders Krogh,et al. Introduction to the theory of neural computation , 1994, The advanced book program.

[10] Thomas M. Cover,et al. Geometrical and Statistical Properties of Systems of Linear Inequalities with Applications in Pattern Recognition , 1965, IEEE Trans. Electron. Comput..

[11] Blatt,et al. Computational capabilities of restricted two-layered perceptrons. , 1994, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[12] E. Gardner. The space of interactions in neural network models , 1988 .

[13] Opper. Learning times of neural networks: Exact solution for a PERCEPTRON algorithm. , 1988, Physical review. A, General physics.