Enregistré dans:
Détails bibliographiques
Auteurs principaux: Walker, Benjamin, McLeod, Andrew D., Qin, Tiexin, Cheng, Yichuan, Li, Haoliang, Lyons, Terry
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2402.18512
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909864229339136
author Walker, Benjamin
McLeod, Andrew D.
Qin, Tiexin
Cheng, Yichuan
Li, Haoliang
Lyons, Terry
author_facet Walker, Benjamin
McLeod, Andrew D.
Qin, Tiexin
Cheng, Yichuan
Li, Haoliang
Lyons, Terry
contents The vector field of a controlled differential equation (CDE) describes the relationship between a control path and the evolution of a solution path. Neural CDEs (NCDEs) treat time series data as observations from a control path, parameterise a CDE's vector field using a neural network, and use the solution path as a continuously evolving hidden state. As their formulation makes them robust to irregular sampling rates, NCDEs are a powerful approach for modelling real-world data. Building on neural rough differential equations (NRDEs), we introduce Log-NCDEs, a novel, effective, and efficient method for training NCDEs. The core component of Log-NCDEs is the Log-ODE method, a tool from the study of rough paths for approximating a CDE's solution. Log-NCDEs are shown to outperform NCDEs, NRDEs, the linear recurrent unit, S5, and MAMBA on a range of multivariate time series datasets with up to $50{,}000$ observations.
format Preprint
id arxiv_https___arxiv_org_abs_2402_18512
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Log Neural Controlled Differential Equations: The Lie Brackets Make a Difference
Walker, Benjamin
McLeod, Andrew D.
Qin, Tiexin
Cheng, Yichuan
Li, Haoliang
Lyons, Terry
Machine Learning
The vector field of a controlled differential equation (CDE) describes the relationship between a control path and the evolution of a solution path. Neural CDEs (NCDEs) treat time series data as observations from a control path, parameterise a CDE's vector field using a neural network, and use the solution path as a continuously evolving hidden state. As their formulation makes them robust to irregular sampling rates, NCDEs are a powerful approach for modelling real-world data. Building on neural rough differential equations (NRDEs), we introduce Log-NCDEs, a novel, effective, and efficient method for training NCDEs. The core component of Log-NCDEs is the Log-ODE method, a tool from the study of rough paths for approximating a CDE's solution. Log-NCDEs are shown to outperform NCDEs, NRDEs, the linear recurrent unit, S5, and MAMBA on a range of multivariate time series datasets with up to $50{,}000$ observations.
title Log Neural Controlled Differential Equations: The Lie Brackets Make a Difference
topic Machine Learning
url https://arxiv.org/abs/2402.18512