Active Sampling Count Sketch (ASCS) for Online Sparse Estimation of a Trillion Scale Covariance Matrix

TL;DR

This paper introduces Active Sampling Count Sketch (ASCS), an online algorithm that efficiently estimates large sparse covariance matrices of trillions of entries, improving accuracy over traditional methods by enhancing the signal-to-noise ratio.

Contribution

The paper presents a novel active sampling strategy integrated with Count Sketch for improved online sparse covariance estimation in high-dimensional data.

Findings

ASCS accurately recovers large covariance entries.

ASCS outperforms vanilla Count Sketch in noisy, high-dimensional settings.

Algorithm is effective on synthetic and real-world datasets.

Abstract

Estimating and storing the covariance (or correlation) matrix of high-dimensional data is computationally challenging because both memory and computational requirements scale quadratically with the dimension. Fortunately, high-dimensional covariance matrices as observed in text, click-through, meta-genomics datasets, etc are often sparse. In this paper, we consider the problem of efficient sparse estimation of covariance matrices with possibly trillions of entries. The size of the datasets we target requires the algorithm to be online, as more than one pass over the data is prohibitive. In this paper, we propose Active Sampling Count Sketch (ASCS), an online and one-pass sketching algorithm, that recovers the large entries of the covariance matrix accurately. Count Sketch (CS), and other sub-linear compressed sensing algorithms, offer a natural solution to the problem in theory. However, vanilla CS does not work well in practice due to a low signal-to-noise ratio (SNR). At the heart of our approach is a novel active sampling strategy that increases the SNR of classical CS. We demonstrate the practicality of our algorithm with synthetic data and real-world high dimensional datasets. ASCS significantly improves over vanilla CS, demonstrating the merit of our active sampling strategy.