Simplified neuron model as a principal component analyzer

@article{Oja1982SimplifiedNM,
  title={Simplified neuron model as a principal component analyzer},
  author={Erkki Oja},
  journal={Journal of Mathematical Biology},
  year={1982},
  volume={15},
  pages={267-273}
}
  • E. Oja
  • Published 1982
  • Biology
  • Journal of Mathematical Biology
A simple linear neuron model with constrained Hebbian-type synaptic modification is analyzed and a new class of unconstrained learning rules is derived. It is shown that the model neuron tends to extract the principal component from a stationary input vector sequence. 

A new simple /spl infin/OH neuron model as a principal component analyzer

  • M. Jankovic
  • Computer Science
    Canadian Conference on Electrical and Computer Engineering 2001. Conference Proceedings (Cat. No.01TH8555)
  • 2001
TLDR
An algorithm for unsupervised learning based on the Hebbian learning rule is presented and a simple "almost linear" neuron model is analyzed, showing that the model neuron tends to extract the principal component from a stationary input vector sequence.

A simple biologically inspired principal component analyzer-ModH neuron model

  • M. Jankovic
  • Computer Science
    6th Seminar on Neural Network Applications in Electrical Engineering
  • 2002
TLDR
The solution proposed here is a modified Hebbian rule in which the modification of the synaptic strength is proportional not to pre- and post-synaptic activity, but instead to the pre- synaptic and averaged value of post- Synaptic activity.

A new simple ∞OH neuron model as a biologically plausible principal component analyzer

  • M. Jankovic
  • Computer Science
    IEEE Trans. Neural Networks
  • 2003
TLDR
The solution proposed here is a modified Hebbian rule, in which the modification of the synaptic strength is proportional not to pre- and post Synaptic activity, but instead to the presynaptic and averaged value of postsynaptic activity.

Study of a Self-Learning Artifical Neuron Model

Giving a large autonomy to each cell and using a statistical average of the entries as a self-learning mechanism are not new ideas [1,2]. But linking statistical and linear dependencies through

A simple, homogeneous parallel PCA network

  • C. Fyfe
  • Computer Science
    Proceedings of 1993 International Conference on Neural Networks (IJCNN-93-Nagoya, Japan)
  • 1993
TLDR
A form of ANN using interneurons has been shown to be capable of performing a principal component analysis of the input data and a new parallel algorithm is proposed using the innate properties of the network.

A Symmetrical Lateral Inhibition Network for PCA and Feature Decorrelation

TLDR
A new network model for data decorrelation and principal component analysis which relays on the biological plausible lateral inhibition and the assignment of the eigenvectors to the neuronal weights are not predefined.

Neural processing of information

  • R. L. Fry
  • Computer Science
    Proceedings of 1994 Workshop on Information Theory and Statistics
  • 1994
TLDR
An application of the Shannon information measures of entropy and mutual information taken together in the context of the proposed model lead to the Hopfield neuron model with a conditionalized Hebbian learning rule and sigmoidal transfer characteristic.

Neural processing of information

  • R. L. Fry
  • Computer Science
    Proceedings of 1994 IEEE International Symposium on Information Theory
  • 1994
TLDR
An application of the Shannon information measures of entropy and mutual information taken together in the context of the proposed model lead to the Hopfield neuron model with a conditionalized Hebbian learning rule and sigmoidal transfer characteristic.

Unsupervised learning algorithm for signal separation

  • T. JacobW. Snyder
  • Computer Science
    2014 IEEE Symposium on Computational Intelligence for Multimedia, Signal and Vision Processing (CIMSIVP)
  • 2014
TLDR
A neural network capable of separating inputs in an unsupervised manner and can operate as a general clustering algorithm, with neighboring neurons responding to geometrically similar inputs is presented.
...

References

SHOWING 1-8 OF 8 REFERENCES

wchastic. approximation methods for constrained and unconstrained systems

TLDR
The Robbins-Monro and Kiefer-Wolfowitz Algorithm for Inequality Constraints and the Weak Convergence of Probability Measures, a simple example, and the Convergence Theorem, a proof of the Main Theorem.

An Introduction to Multivariate Statistical Analysis

Preface to the Third Edition.Preface to the Second Edition.Preface to the First Edition.1. Introduction.2. The Multivariate Normal Distribution.3. Estimation of the Mean Vector and the Covariance

Development of Specificity in the Cat Visual Cortex

A physiological mechanism for Hebb's postulate of learning.

  • G. Stent
  • Biology
    Proceedings of the National Academy of Sciences of the United States of America
  • 1973
TLDR
The physiological mechanism proposed here for this process posits that at synapses acting according to Hebb's postulate, the receptors for the neurotransmitter are eliminated from the post Synaptic membrane by the transient reversals of the sign of membrane polarization that occur during action potential impulses in the postsynaptic cell.

On stochastic approximation of eigenvectors and eigenvalues of the expectation

  • 1975