MTS-JEPA: Multi-Resolution Joint-Embedding Predictive Architecture for Time-Series Anomaly Prediction

TL;DR

MTS-JEPA is a novel multi-resolution architecture for time-series anomaly prediction that effectively captures transient shocks and long-term trends, preventing collapse and achieving state-of-the-art results.

Contribution

It introduces a multi-resolution predictive objective with a soft codebook to improve representation learning and anomaly detection in multivariate time series.

Findings

Prevents representation collapse in JEPA models.

Achieves state-of-the-art early-warning anomaly detection performance.

Effectively captures regime transitions and long-term trends.

Abstract

Multivariate time series underpin modern critical infrastructure, making the prediction of anomalies a vital necessity for proactive risk mitigation. While Joint-Embedding Predictive Architectures (JEPA) offer a promising framework for modeling the latent evolution of these systems, their application is hindered by representation collapse and an inability to capture precursor signals across varying temporal scales. To address these limitations, we propose MTS-JEPA, a specialized architecture that integrates a multi-resolution predictive objective with a soft codebook bottleneck. This design explicitly decouples transient shocks from long-term trends, and utilizes the codebook to capture discrete regime transitions. Notably, we find this constraint also acts as an intrinsic regularizer to ensure optimization stability. Empirical evaluations on standard benchmarks confirm that our approach effectively prevents degenerate solutions and achieves state-of-the-art performance under the early-warning protocol.