OATS: Online Data Augmentation for Time Series Foundation Models

TL;DR

OATS introduces a dynamic, sample-guided data augmentation method for Time Series Foundation Models, significantly improving performance by tailoring synthetic data generation to training stages using a diffusion-based approach.

Contribution

The paper presents OATS, a novel online data augmentation strategy that adaptively generates synthetic time series data conditioned on training samples and stages, outperforming static methods.

Findings

OATS consistently improves model performance over static augmentation methods.

The diffusion-based framework produces realistic and diverse synthetic time series.

OATS achieves substantial gains across multiple datasets and architectures.

Abstract

Time Series Foundation Models (TSFMs) are a powerful paradigm for time series analysis and are often enhanced by synthetic data augmentation to improve the training data quality. Existing augmentation methods, however, typically rely on heuristics and static paradigms. Motivated by dynamic data optimization, which shows that the contribution of samples varies across training stages, we propose OATS (Online Data Augmentation for Time Series Foundation Models), a principled strategy that generates synthetic data tailored to different training steps. OATS leverages valuable training samples as principled guiding signals and dynamically generates high-quality synthetic data conditioned on them. We further design a diffusion-based framework to produce realistic time series and introduce an explore-exploit mechanism to balance efficiency and effectiveness. Experiments on TSFMs demonstrate that OATS consistently outperforms regular training and yields substantial performance gains over static data augmentation baselines across six validation datasets and two TSFM architectures. The code is available at the link https://github.com/microsoft/TimeCraft.