Expected signature of Gaussian processes with strictly regular kernels

TL;DR

This paper calculates the expected signature of a specific subclass of Gaussian processes with regular kernels, providing insights into their mathematical structure.

Contribution

It introduces a method to compute the expected signature for Gaussian processes with strictly regular kernels, expanding understanding of their properties.

Findings

Derived explicit formulas for expected signatures

Enhanced understanding of Gaussian processes with regular kernels

Potential applications in stochastic analysis and machine learning

Abstract

We compute the expected signature of a class of Gaussian processes which is a subclass of the Gaussian processes with regular kernels, in the sense of Alos, Mazet and Nualart.