Generation of unpredictable time series by a Neural Network

Abstract

A perceptron that learns the opposite of its own output is used to generate a time series. We analyse properties of the weight vector and the generated sequence, like the cycle length and the probability distribution of generated sequences. A remarkable suppression of the autocorrelation function is explained, and connections to the Bernasconi model are discussed. If a continuous transfer function is used, the system displays chaotic and intermittent behaviour, with the product of the learning rate and amplification as a control parameter.

11 pages, 14 figures; slightly expanded and clarified, mistakes corrected; accepted for publication in PRE

References in corpus (4)

• Error-Correcting Codes That Nearly Saturate Shannon's Bound
• Self-Averaging and On-line Learning
• Long-term properties of time series generated by a perceptron with various transfer functions
• Noisy time series generation by feed-forward networks

Cited by in corpus (5)

• Neural cryptography with feedback
• Neural networks with transient state dynamics
• Neural Synchronization and Cryptography
• Antipersistent binary time series
• Neural Networks, Game Theory and Time Series Generation