TimeRadar: A Domain-Rotatable Foundation Model for Time Series Anomaly Detection

2026-02-22

AI Summary

This work addresses the limitation of existing time series foundation models, which are confined to predefined time-frequency domains and struggle to capture rare, irregular patterns in unsupervised anomaly detection. To overcome this, we propose a novel foundation model capable of adaptive rotation in fractional-order time-frequency domains. By introducing a learnable fractional-order time-frequency rotation mechanism, the model dynamically optimizes its representation angle to effectively discriminate between normal and anomalous instances. Our approach integrates Fractional-order Time-Frequency Reconstruction (FTFRecon) with Contextual Deviation Learning (CDL), enabling unified modeling of global reconstruction and local contextual deviations. This is the first domain-rotatable foundation model tailored for unsupervised anomaly detection. Experiments demonstrate that our method significantly improves cross-domain anomaly detection performance on both seen and unseen datasets, effectively identifying unbounded anomalous patterns.

Technology Category

• Machine Learning: Other Foundations of Machine Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Computer Vision: Other Foundations of Computer Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Current time series foundation models (TSFMs) primarily focus on learning prevalent and regular patterns within a predefined time or frequency domain to enable supervised downstream tasks (e.g., forecasting). Consequently, they are often ineffective for inherently unsupervised downstream tasks-such as time series anomaly detection (TSAD), which aims to identify rare, irregular patterns. This limitation arises because such abnormal patterns can closely resemble the regular patterns when presented in the same time/frequency domain. To address this issue, we introduce TimeRadar, an innovative TSFM built in a fractional time-frequency domain to support generalist TSAD across diverse unseen datasets. Our key insight is that rotating a time series into a data-dependent fractional time-frequency representation can adaptively differentiate the normal and abnormal signals across different datasets. To this end, a novel component, namely Fractionally modulated Time-Frequency Reconstruction (FTFRecon), is proposed in TimeRadar to leverage a learnable fractional order to rotate the time series to the most pronounced angle between a continuous time and frequency domain for accurate data reconstruction. This provides adaptive data reconstruction in an optimal time-frequency domain for each data input, enabling effective differentiation of the unbounded abnormal patterns from the regular ones across datasets, including unseen datasets. To allow TimeRadar to model local abnormality that is not captured by the global data reconstruction, we further introduce a Contextual Deviation Learning (CDL) component to model the local deviation of the input relative to its contextual time series data in the rotatable domain.

Problem

• time series anomaly detection
• foundation model
• unsupervised learning
• time-frequency domain
• abnormal pattern

Innovation

• fractional time-frequency domain
• time series anomaly detection
• foundation model
• adaptive reconstruction
• contextual deviation learning