Non-Sparse PCA in High Dimensions via Cone Projected Power Iteration

TL;DR

This paper introduces a cone projected power iteration method for efficiently recovering the principal eigenvector in high-dimensional noisy settings, especially when the eigenvector belongs to a convex cone, improving speed and accuracy.

Contribution

The paper presents a novel cone projected power iteration algorithm with polynomial time complexity for convex cones, providing theoretical error bounds and demonstrating superior performance over existing methods.

Findings

Achieves faster computation than traditional power iteration.

Attains smaller error when the matrix has low cone-restricted operator norm.

Outperforms sparse PCA algorithms in relevant scenarios.

Abstract

In this paper, we propose a cone projected power iteration algorithm to recover the first principal eigenvector from a noisy positive semidefinite matrix. When the true principal eigenvector is assumed to belong to a convex cone, the proposed algorithm is fast and has a tractable error. Specifically, the method achieves polynomial time complexity for certain convex cones equipped with fast projection such as the monotone cone. It attains a small error when the noisy matrix has a small cone-restricted operator norm. We supplement the above results with a minimax lower bound of the error under the spiked covariance model. Our numerical experiments on simulated and real data, show that our method achieves shorter run time and smaller error in comparison to the ordinary power iteration and some sparse principal component analysis algorithms if the principal eigenvector is in a convex cone.