Tuning-free one-bit covariance estimation using data-driven dithering

TL;DR

This paper introduces a data-driven, tuning-free method for one-bit covariance estimation of subgaussian distributions, achieving near-optimal error bounds without prior knowledge of distribution variances.

Contribution

It proposes a novel, tuning-free estimator that replaces the unknown parameter with a data-driven quantity, maintaining near-minimax optimal error bounds.

Findings

The estimator achieves near-minimax optimal error bounds.

Refined dithering allows estimation error comparable to the unquantized sample covariance.

The method is tuning-free and practical for real-world applications.

Abstract

We consider covariance estimation of any subgaussian distribution from finitely many i.i.d. samples that are quantized to one bit of information per entry. Recent work has shown that a reliable estimator can be constructed if uniformly distributed dithers on $[-\lambda,\lambda]$ are used in the one-bit quantizer. This estimator enjoys near-minimax optimal, non-asymptotic error estimates in the operator and Frobenius norms if $\lambda$ is chosen proportional to the largest variance of the distribution. However, this quantity is not known a-priori, and in practice $\lambda$ needs to be carefully tuned to achieve good performance. In this work we resolve this problem by introducing a tuning-free variant of this estimator, which replaces $\lambda$ by a data-driven quantity. We prove that this estimator satisfies the same non-asymptotic error estimates - up to small (logarithmic) losses and a slightly worse probability estimate. We also show that by using refined data-driven dithers that vary per entry of each sample, one can construct an estimator satisfying the same estimation error bound as the sample covariance of the samples before quantization -- again up logarithmic losses. Our proofs rely on a new version of the Burkholder-Rosenthal inequalities for matrix martingales, which is expected to be of independent interest.