AdaNODEs: Test Time Adaptation for Time Series Forecasting Using Neural ODEs

2026-01-19

AI Summary

This work addresses the challenge of test-time distribution shift in time series forecasting, where existing test-time adaptation (TTA) methods—often assuming independent and identically distributed data—struggle to perform effectively. To overcome this limitation, the authors propose AdaNODEs, the first method to integrate Neural Ordinary Differential Equations (Neural ODEs) into a source-free TTA framework tailored for time series. AdaNODEs introduces a lightweight adaptation mechanism that models continuous-time dynamics and employs a novel loss function, enabling efficient adaptation to distribution shifts by updating only a small subset of parameters. Experimental results demonstrate that AdaNODEs achieves relative performance improvements of 5.88% and 28.4% on univariate and multivariate time series benchmarks, respectively, significantly outperforming current state-of-the-art approaches and exhibiting robustness under severe distribution shifts.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Search and Optimization: Distributed Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

Test time adaptation (TTA) has emerged as a promising solution to adapt pre-trained models to new, unseen data distributions using unlabeled target domain data. However, most TTA methods are designed for independent data, often overlooking the time series data and rarely addressing forecasting tasks. This paper presents AdaNODEs, an innovative source-free TTA method tailored explicitly for time series forecasting. By leveraging Neural Ordinary Differential Equations (NODEs), we propose a novel adaptation framework that accommodates the unique characteristics of distribution shifts in time series data. Moreover, we innovatively propose a new loss function to tackle TTA for forecasting tasks. AdaNODEs only requires updating limited model parameters, showing effectiveness in capturing temporal dependencies while avoiding significant memory usage. Extensive experiments with one- and high-dimensional data demonstrate that AdaNODEs offer relative improvements of 5.88\% and 28.4\% over the SOTA baselines, especially demonstrating robustness across higher severity distribution shifts.

Problem

• Test Time Adaptation
• Time Series Forecasting
• Distribution Shift
• Neural ODEs
• Source-Free Adaptation

Innovation

• Test Time Adaptation
• Neural ODEs
• Time Series Forecasting
• Source-Free Adaptation
• Distribution Shift