General Signature Kernels

DOI10.1214/23-AAP1973MaRDI QIDQ6504561arXivFDO

Authors Thomas Cass, Terence J. Lyons, Xingcheng Xu

Published in arXiv

Full work available at URL https://doi.org/10.1214/23-AAP1973

Mathematics Subject Classification ID

Signatures and data streams (60L10) Applications of rough analysis (60L90) Parameterized complexity, tractability and kernelization (68Q27)

Abstract: Suppose that $gamma$ and $sigma$ are two continuous bounded variation paths which take values in a finite-dimensional inner product space $V$. Recent papers have introduced the truncated and the untruncated signature kernel of $gamma$ and $sigma$, and showed how these concepts can be used in classification and prediction tasks involving multivariate time series. In this paper, we introduce general signature kernels and show how they can be interpreted, in many examples, as an average of PDE solutions, and hence how they can be computed using suitable quadrature rules. We extend this analysis to derive closed-form formulae for expressions involving the expected (Stratonovich) signature of Brownian motion. In doing so, we articulate a novel connection between signature kernels and the hyperbolic development map, the latter of which has been a broadly useful tool in the analysis of the signature. As an application we evaluate the use of different general signature kernels as the basis for non-parametric goodness-of-fit tests to Wiener measure on path space.

This page was built for publication: General Signature Kernels

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6504561)