Training a perceptron by a bit sequence: storage capacity

DOI10.1088/0305-4470/29/24/020MaRDI QIDQ4394424zbMATH OpenOpenAlexFDO

Authors

Publication date 10 September 1998

Published in Journal of Physics A: Mathematical and General (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/cond-mat/9607040

zbMATH Keywords

classification problemperceptron

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Neural nets applied to problems in time-dependent statistical mechanics (82C32)

Abstract: A perceptron is trained by a random bit sequence. In comparison to the corresponding classification problem, the storage capacity decreases to alpha_c=1.70pm 0.02 due to correlations between input and output bits. The numerical results are supported by a signal to noise analysis of Hebbian weights.

Recommendations

• Storage of correlated patterns in a perceptron
• Information capacity of a perceptron
• Capacity of the single-layer perceptron and minimal trajectory training algorithms
• Storage capacity and optimal learning of Potts-model perceptrons by a cavity method

Cited in

(3)

• Information capacity of a perceptron
• Capacity of the single-layer perceptron and minimal trajectory training algorithms
• Statistical physics of interacting neural networks

This page was built for publication: Training a perceptron by a bit sequence: storage capacity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4394424)