A Self-consistent-field Iteration for Orthogonal Canonical Correlation Analysis

TL;DR

This paper introduces a novel, efficient self-consistent-field iteration algorithm for orthogonal canonical correlation analysis (OCCA), ensuring convergence and improving performance over existing methods, with applications in multi-label classification and multi-view feature extraction.

Contribution

It presents a new SCF-based iterative algorithm for OCCA that guarantees convergence and outperforms existing methods in efficiency and accuracy.

Findings

The proposed algorithm converges globally to a KKT point.

It outperforms baseline methods in experiments.

It is effective in real-world applications like multi-label classification.

Abstract

We propose an efficient algorithm for solving orthogonal canonical correlation analysis (OCCA) in the form of trace-fractional structure and orthogonal linear projections. Even though orthogonality has been widely used and proved to be a useful criterion for pattern recognition and feature extraction, existing methods for solving OCCA problem are either numerical unstable by relying on a deflation scheme, or less efficient by directly using generic optimization methods. In this paper, we propose an alternating numerical scheme whose core is the sub-maximization problem in the trace-fractional form with an orthogonal constraint. A customized self-consistent-field (SCF) iteration for this sub-maximization problem is devised. It is proved that the SCF iteration is globally convergent to a KKT point and that the alternating numerical scheme always converges. We further formulate a new trace-fractional maximization problem for orthogonal multiset CCA (OMCCA) and then propose an efficient algorithm with an either Jacobi-style or Gauss-Seidel-style updating scheme based on the same SCF iteration. Extensive experiments are conducted to evaluate the proposed algorithms against existing methods including two real world applications: multi-label classification and multi-view feature extraction. Experimental results show that our methods not only perform competitively to or better than baselines but also are more efficient.