Ranking Neighborhood and Class Prototype Contrastive Learning for Time Series

Time series are often complex and rich in information but sparsely labeled and therefore challenging to model. Existing contrastive learning methods conduct augmentations and maximize their similarity. However, they ignore the similarity of adjacent timestamps and suffer from the problem of sampling bias. In this paper, we propose a self-supervised framework for learning generalizable representations of time series, called <inline-formula><tex-math notation="LaTeX">$\mathbf {R}$</tex-math></inline-formula>anking n<inline-formula><tex-math notation="LaTeX">$\mathbf {E}$</tex-math></inline-formula> ighborhood and cla<inline-formula><tex-math notation="LaTeX">$\mathbf {S}$</tex-math></inline-formula>s prototyp<inline-formula><tex-math notation="LaTeX">$\mathbf {E}$</tex-math></inline-formula> contr<inline-formula><tex-math notation="LaTeX">$\mathbf {A}$</tex-math></inline-formula>stive <inline-formula><tex-math notation="LaTeX">$\mathbf {L}$</tex-math></inline-formula>earning (RESEAL). It exploits information about similarity ranking to learn an embedding space, ensuring that positive samples are ranked according to their temporal order. Additionally, RESEAL introduces a class prototype contrastive learning module. It contrasts time series representations and their corresponding centroids as positives against truly negative pairs from different clusters, mitigating the sampling bias issue. Extensive experiments conducted on several multivariate and univariate time series tasks (i.e., classification, anomaly detection, and forecasting) demonstrate that our representation framework achieves significant improvement over existing baselines of self-supervised time series representation.