A geometric newton method for oja's vector field

Authors:
P. -A. Absil;M. Ishteva;L. De Lathauwer;S. Van Huffel
Affiliations:
-;-;-;-
Venue:
Neural Computation
Year:
2009

Citing 14
Cited 4

Original Contribution: Least mean square error reconstruction principle for self-organizing neural-nets

Neural Networks
Choosing the forcing terms in an inexact Newton method

SIAM Journal on Scientific Computing - Special issue on iterative methods in numerical linear algebra; selected papers from the Colorado conference
Matrix computations (3rd ed.)

Matrix computations (3rd ed.)
Stability and Convergence of Principal Component Learning Algorithms

SIAM Journal on Matrix Analysis and Applications
The Geometry of Algorithms with Orthogonality Constraints

SIAM Journal on Matrix Analysis and Applications
Iterative solution of nonlinear equations in several variables

Iterative solution of nonlinear equations in several variables
Trust Region Algorithms and Timestep Selection

SIAM Journal on Numerical Analysis
Unified stabilization approach to principal and minor components extraction algorithms

Neural Networks
Adaptive Eigenvalue Computations Using Newton's Method on the Grassmann Manifold

SIAM Journal on Matrix Analysis and Applications
Cubically Convergent Iterations for Invariant Subspace Computation

SIAM Journal on Matrix Analysis and Applications
Pseudo-Transient Continuation for Nonsmooth Nonlinear Equations

SIAM Journal on Numerical Analysis
Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)

Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)
Optimization Algorithms on Matrix Manifolds

Optimization Algorithms on Matrix Manifolds
Modulated Hebb-Oja learning Rule-a method for principal subspace analysis

IEEE Transactions on Neural Networks

Riemannian Metric and Geometric Mean for Positive Semidefinite Matrices of Fixed Rank

SIAM Journal on Matrix Analysis and Applications
Low-Rank Optimization on the Cone of Positive Semidefinite Matrices

SIAM Journal on Optimization
A Riemannian Optimization Approach for Computing Low-Rank Solutions of Lyapunov Equations

SIAM Journal on Matrix Analysis and Applications
Best Low Multilinear Rank Approximation of Higher-Order Tensors, Based on the Riemannian Trust-Region Scheme

SIAM Journal on Matrix Analysis and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Newton's method for solving the matrix equation runs up against the fact that its zeros are not isolated. This is due to a symmetry of F by the action of the orthogonal group. We show how differential-geometric techniques can be exploited to remove this symmetry and obtain a “geometric” Newton algorithm that finds the zeros of F. The geometric Newton method does not suffer from the degeneracy issue that stands in the way of the original Newton method.