ML Anthology

Time-Domain Solutions of Oja's Equations

John L. Wyatt Jr., Ibrahim M. Elfadel
NeCo 1995 pp. 915-922
doi:10.1162/NECO.1995.7.5.915 /neco/1995/jr1995neco-timedomain/

Abstract

Oja's equations describe a well-studied system for unsupervised Hebbian learning of principal components. This paper derives the explicit time-domain solution of Oja's equations for the single-neuron case. It also shows that, under a linear change of coordinates, these equations are a gradient system in the general multi-neuron case. This latter result leads to a new Lyapunov-like function for Oja's equations.

NeCo Semantic Scholar

Cite

Text

Jr. and Elfadel. "Time-Domain Solutions of Oja's Equations." Neural Computation, 1995. doi:10.1162/NECO.1995.7.5.915

Markdown

[Jr. and Elfadel. "Time-Domain Solutions of Oja's Equations." Neural Computation, 1995.](https://mlanthology.org/neco/1995/jr1995neco-timedomain/) doi:10.1162/NECO.1995.7.5.915

BibTeX

@article{jr1995neco-timedomain,
  title     = {{Time-Domain Solutions of Oja's Equations}},
  author    = {Jr., John L. Wyatt and Elfadel, Ibrahim M.},
  journal   = {Neural Computation},
  year      = {1995},
  pages     = {915-922},
  doi       = {10.1162/NECO.1995.7.5.915},
  volume    = {7},
  url       = {https://mlanthology.org/neco/1995/jr1995neco-timedomain/}
}