LDTC: Lifelong deep temporal clustering for multivariate time series

TL;DR

LDTC introduces a lifelong deep learning framework for multivariate time series clustering, capable of continuously learning from dynamic data without forgetting previous knowledge, outperforming existing methods.

Contribution

The paper presents LDTC, a novel lifelong deep temporal clustering algorithm that integrates dimensionality reduction and dynamic model expansion for sequential time series tasks.

Findings

Achieves high-quality clustering on seven real-world datasets.

Effectively learns new tasks without catastrophic forgetting.

Outperforms existing temporal clustering methods.

Abstract

Clustering temporal and dynamically changing multivariate time series from real-world fields, called temporal clustering for short, has been a major challenge due to inherent complexities. Although several deep temporal clustering algorithms have demonstrated a strong advantage over traditional methods in terms of model learning and clustering results, the accuracy of the few algorithms are not satisfactory. None of the existing algorithms can continuously learn new tasks and deal with the dynamic data effectively and efficiently in the sequential tasks learning. To bridge the gap and tackle these issues, this paper proposes a novel algorithm \textbf{L}ifelong \textbf{D}eep \textbf{T}emporal \textbf{C}lustering (\textbf{LDTC}), which effectively integrates dimensionality reduction and temporal clustering into an end-to-end deep unsupervised learning framework. Using a specifically designed autoencoder and jointly optimizing for both the latent representation and clustering objective, the LDTC can achieve high-quality clustering results. Moreover, unlike any previous work, the LDTC is uniquely equipped with the fully dynamic model expansion and rehearsal-based techniques to effectively learn new tasks and to tackle the dynamic data in the sequential tasks learning without the catastrophic forgetting or degradation of the model accuracy. Experiments on seven real-world multivariate time series datasets show that the LDTC is a promising method for dealing with temporal clustering issues effectively and efficiently.