Learning with Expected Signatures: Theory and Applications

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lucchese, Lorenzo, Pakkanen, Mikko S., Veraart, Almut E. D.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908382610325504
author Lucchese, Lorenzo
Pakkanen, Mikko S.
Veraart, Almut E. D.
author_facet Lucchese, Lorenzo
Pakkanen, Mikko S.
Veraart, Almut E. D.
contents The expected signature maps a collection of data streams to a lower dimensional representation, with a remarkable property: the resulting feature tensor can fully characterize the data generating distribution. This "model-free" embedding has been successfully leveraged to build multiple domain-agnostic machine learning (ML) algorithms for time series and sequential data. The convergence results proved in this paper bridge the gap between the expected signature's empirical discrete-time estimator and its theoretical continuous-time value, allowing for a more complete probabilistic interpretation of expected signature-based ML methods. Moreover, when the data generating process is a martingale, we suggest a simple modification of the expected signature estimator with significantly lower mean squared error and empirically demonstrate how it can be effectively applied to improve predictive performance.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20465
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning with Expected Signatures: Theory and Applications
Lucchese, Lorenzo
Pakkanen, Mikko S.
Veraart, Almut E. D.
Machine Learning
Probability
Statistics Theory
The expected signature maps a collection of data streams to a lower dimensional representation, with a remarkable property: the resulting feature tensor can fully characterize the data generating distribution. This "model-free" embedding has been successfully leveraged to build multiple domain-agnostic machine learning (ML) algorithms for time series and sequential data. The convergence results proved in this paper bridge the gap between the expected signature's empirical discrete-time estimator and its theoretical continuous-time value, allowing for a more complete probabilistic interpretation of expected signature-based ML methods. Moreover, when the data generating process is a martingale, we suggest a simple modification of the expected signature estimator with significantly lower mean squared error and empirically demonstrate how it can be effectively applied to improve predictive performance.
title Learning with Expected Signatures: Theory and Applications
topic Machine Learning
Probability
Statistics Theory
url https://arxiv.org/abs/2505.20465