A network for recursive extraction of canonical coordinates

Authors:
Ali Pezeshki;Mahmood R. Azimi-Sadjadi;Louis L. Scharf
Affiliations:
Department of Electrical and Computer Engineering, Colorado State University, Fort Collins, CO;Department of Electrical and Computer Engineering, Colorado State University, Fort Collins, CO;Department of Electrical and Computer Engineering, Colorado State University, Fort Collins, CO
Venue:
Neural Networks - 2003 Special issue: Advances in neural networks research — IJCNN'03
Year:
2003

Citing 7
Cited 4

Adaptive filter theory (3rd ed.)

Adaptive filter theory (3rd ed.)
Principal component neural networks: theory and applications

Principal component neural networks: theory and applications
A neural implementation of canonical correlation analysis

Neural Networks
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Canonical coordinates and the geometry of inference, rate, andcapacity

IEEE Transactions on Signal Processing
Wiener filters in canonical coordinates for transform coding,filtering, and quantizing

IEEE Transactions on Signal Processing
Principal component extraction using recursive least squares learning

IEEE Transactions on Neural Networks

A learning algorithm for adaptive canonical correlation analysis of several data sets

Neural Networks
2008 Special Issue: Iterative multi-channel coherence analysis with applications

Neural Networks
Joint blind source separation by multiset canonical correlation analysis

IEEE Transactions on Signal Processing
Adaptive reduced-rank constrained constant modulus algorithms based on joint iterative optimization of filters for beamforming

IEEE Transactions on Signal Processing

Quantified Score

Hi-index	0.01

Visualization

Abstract

A network structure for canonical coordinate decomposition is presented. The network consists of two single-layer linear subnetworks that together extract the canonical coordinates of two data channels. The connection weights of the networks are trained by a stochastic gradient descent learning algorithm. Each subnetwork features a hierarchical set of lateral connections among its outputs. The lateral connections perform a deflation process that subtracts the contributions of the already extracted coordinates from the input data subspace. This structure allows for adding new nodes for extracting additional canonical coordinates without the need for retraining the previous nodes. The performance of the network is evaluated on a synthesized data set.