Streaming PCA for Markovian Data

Syamantak Kumar, Purnamrita Sarkar

Keywords: Streaming PCA, Markov Chain, Mixing, Oja's algorithm

TL;DR: We prove near-optimal rate of convergence of Oja's Streaming PCA algorithm for data sampled from a Markov chain and show that it is better than discarding data to reduce dependence at finding the leading eigenvector.

Abstract: Since its inception in 1982, Oja's algorithm has become an established method for streaming principle component analysis (PCA). We study the problem of streaming PCA, where the data-points are sampled from an irreducible, aperiodic, and reversible Markov chain starting in stationarity. Our goal is to estimate the top eigenvector of the unknown covariance matrix of the stationary distribution. This setting has implications in scenarios where data can solely be sampled from a Markov Chain Monte Carlo (MCMC) type algorithm, and the objective is to perform inference on parameters of the stationary distribution. Most convergence guarantees for Oja's algorithm in the literature assume that the data-points are sampled IID. For data streams with Markovian dependence, one typically downsamples the data to get a "nearly" independent data stream. In this paper, we obtain the first near-optimal rate for Oja's algorithm on the entire data, where we remove the logarithmic dependence on the sample size, $n$, resulting from throwing data away in downsampling strategies.

Submission Number: 12013