Doubly Accelerated Methods for Faster CCA and Generalized Eigendecomposition

Abstract

We study $k$-GenEV, the problem of finding the top $k$ generalized eigenvectors, and $k$-CCA, the problem of finding the top $k$ vectors in canonical-correlation analysis. We propose algorithms $\mathtt{LazyEV}$ and $\mathtt{LazyCCA}$ to solve the two problems with running times linearly dependent on the input size and on $k$. Furthermore, our algorithms are DOUBLY-ACCELERATED: our running times depend only on the square root of the matrix condition number, and on the square root of the eigengap. This is the first such result for both $k$-GenEV or $k$-CCA. We also provide the first gap-free results, which provide running times that depend on $1/\sqrt{\epsilon }$ rather than the eigengap.

We have now stated more clearly why this paper has outperformed relevant previous results, and included discussions for doubly-stochastic methods. arXiv admin note: text overlap with arXiv:1607.03463