TS2Vec-Ensemble: An Enhanced Self-Supervised Framework for Time Series Forecasting
This work solves the problem of inaccurate long-horizon forecasting in time series analysis for researchers and practitioners, though it is incremental as it builds upon existing TS2Vec methods.
The paper tackles the problem of time series forecasting by addressing the limitations of self-supervised models like TS2Vec, which prioritize instance discrimination over capturing deterministic patterns like seasonality and trend, and introduces TS2Vec-Ensemble, a hybrid framework that fuses learned dynamics with explicit time features, resulting in consistent and significant performance improvements over baselines on ETT benchmark datasets.
Self-supervised representation learning, particularly through contrastive methods like TS2Vec, has advanced the analysis of time series data. However, these models often falter in forecasting tasks because their objective functions prioritize instance discrimination over capturing the deterministic patterns, such as seasonality and trend, that are critical for accurate prediction. This paper introduces TS2Vec-Ensemble, a novel hybrid framework designed to bridge this gap. Our approach enhances the powerful, implicitly learned dynamics from a pretrained TS2Vec encoder by fusing them with explicit, engineered time features that encode periodic cycles. This fusion is achieved through a dual-model ensemble architecture, where two distinct regression heads -- one focused on learned dynamics and the other on seasonal patterns -- are combined using an adaptive weighting scheme. The ensemble weights are optimized independently for each forecast horizon, allowing the model to dynamically prioritize short-term dynamics or long-term seasonality as needed. We conduct extensive experiments on the ETT benchmark datasets for both univariate and multivariate forecasting. The results demonstrate that TS2Vec-Ensemble consistently and significantly outperforms the standard TS2Vec baseline and other state-of-the-art models, validating our hypothesis that a hybrid of learned representations and explicit temporal priors is a superior strategy for long-horizon time series forecasting.