Neural CDEs as Correctors for Learned Time Series Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shahid, Muhammad Bilal, Jiang, Zhanhong, Koirala, Prajwal, Sarkar, Soumik, Fleming, Cody
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915990263037952
author Shahid, Muhammad Bilal
Jiang, Zhanhong
Koirala, Prajwal
Sarkar, Soumik
Fleming, Cody
author_facet Shahid, Muhammad Bilal
Jiang, Zhanhong
Koirala, Prajwal
Sarkar, Soumik
Fleming, Cody
contents Learned time-series models, whether continuous or discrete, are widely used for forecasting the states of dynamical systems but suffer from error accumulation in multi-step forecasts. To address this issue, we propose a Predictor-Corrector framework in which the Predictor is a learned time-series model that generates multi-step forecasts and the Corrector is a neural controlled differential equation that corrects the forecast errors. The Corrector works with irregularly sampled time series and is compatible with both continuous- and discrete-time Predictors. We further introduce two regularization strategies that improve the Corrector's extrapolation performance and accelerate its training. We also provide theoretical guarantees on the stability and convergence of the proposed framework. Experiments on synthetic, physics-based, and real-world datasets show that the proposed framework consistently improves forecasting performance across diverse Predictors, including neural ordinary differential equations, ContiFormer, and DLinear, demonstrating its predictor-agnostic nature.
format Preprint
id arxiv_https___arxiv_org_abs_2512_12116
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Neural CDEs as Correctors for Learned Time Series Models
Shahid, Muhammad Bilal
Jiang, Zhanhong
Koirala, Prajwal
Sarkar, Soumik
Fleming, Cody
Machine Learning
Learned time-series models, whether continuous or discrete, are widely used for forecasting the states of dynamical systems but suffer from error accumulation in multi-step forecasts. To address this issue, we propose a Predictor-Corrector framework in which the Predictor is a learned time-series model that generates multi-step forecasts and the Corrector is a neural controlled differential equation that corrects the forecast errors. The Corrector works with irregularly sampled time series and is compatible with both continuous- and discrete-time Predictors. We further introduce two regularization strategies that improve the Corrector's extrapolation performance and accelerate its training. We also provide theoretical guarantees on the stability and convergence of the proposed framework. Experiments on synthetic, physics-based, and real-world datasets show that the proposed framework consistently improves forecasting performance across diverse Predictors, including neural ordinary differential equations, ContiFormer, and DLinear, demonstrating its predictor-agnostic nature.
title Neural CDEs as Correctors for Learned Time Series Models
topic Machine Learning
url https://arxiv.org/abs/2512.12116