CovNet: Covariance Networks for Functional Data on Multidimensional Domains

TL;DR

This paper introduces Covariance Networks (CovNet), a flexible and efficient neural network-based method for estimating covariance functions in multidimensional functional data, overcoming computational and statistical challenges.

Contribution

The paper proposes CovNet, a universal neural network model for covariance estimation in multidimensional functional data, with efficient fitting, closed-form eigendecomposition, and proven consistency.

Findings

CovNet can approximate any covariance function with desired accuracy.

The method is computationally efficient and scalable.

Demonstrated effectiveness on simulation and fMRI data.

Abstract

Covariance estimation is ubiquitous in functional data analysis. Yet, the case of functional observations over multidimensional domains introduces computational and statistical challenges, rendering the standard methods effectively inapplicable. To address this problem, we introduce "Covariance Networks" (CovNet) as a modeling and estimation tool. The CovNet model is "universal" - it can be used to approximate any covariance up to desired precision. Moreover, the model can be fitted efficiently to the data and its neural network architecture allows us to employ modern computational tools in the implementation. The CovNet model also admits a closed-form eigendecomposition, which can be computed efficiently, without constructing the covariance itself. This facilitates easy storage and subsequent manipulation of a covariance in the context of the CovNet. We establish consistency of the proposed estimator and derive its rate of convergence. The usefulness of the proposed method is demonstrated by means of an extensive simulation study and an application to resting state fMRI data.