Some Pathological Traps for Stochastic Approximation
Author:
- Odile BrandièreOBOdile Brandière
- 1 July 1998
- journal article
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Control and Optimization
- Vol. 36 (4) , 1293-1314
Abstract
We consider different kinds of "pathological traps" for stochastic algorithms, thus extending a previous study on regular traps. An illustration is given by the complete proof of the convergence of a principal component analysis (PCA) algorithm when the eigenvalues are multiple.Keywords
This publication has 13 references indexed in Scilit:
- General results on the convergence of stochastic algorithmsIEEE Transactions on Automatic Control, 1996
- A Dynamical System Approach to Stochastic ApproximationsSIAM Journal on Control and Optimization, 1996
- Dynamics of Morse-Smale urn processesErgodic Theory and Dynamical Systems, 1995
- Principal components, minor components, and linear neural networksNeural Networks, 1992
- Convergence analysis of local feature extraction algorithmsNeural Networks, 1992
- On stochastic approximation of the eigenvectors and eigenvalues of the expectation of a random matrixJournal of Mathematical Analysis and Applications, 1985
- Simplified neuron model as a principal component analyzerJournal of Mathematical Biology, 1982
- Stochastic Approximation Methods for Constrained and Unconstrained Systems1978
- Approximation stochastique dans une algèbre normée non commutativeBulletin de la Société Mathématiques de France, 1969
- Sur les séries dont les termes sont des variables éventuelles indépendantesStudia Mathematica, 1931