Attention-Guided Deep Adversarial Temporal Subspace Clustering (A-DATSC) Model for multivariate spatiotemporal data

TL;DR

This paper introduces A-DATSC, a novel deep adversarial clustering model that effectively captures complex spatiotemporal dependencies in multivariate data, outperforming existing methods in real-world applications.

Contribution

The paper presents a new attention-guided deep adversarial model with a U-Net inspired generator and graph attention transformer, addressing limitations of shallow autoencoders and enhancing spatiotemporal feature modeling.

Findings

A-DATSC achieves superior clustering accuracy on real-world datasets.

The model effectively captures local and global spatiotemporal dependencies.

Experimental results outperform state-of-the-art deep subspace clustering methods.

Abstract

Deep subspace clustering models are vital for applications such as snowmelt detection, sea ice tracking, crop health monitoring, infectious disease modeling, network load prediction, and land-use planning, where multivariate spatiotemporal data exhibit complex temporal dependencies and reside on multiple nonlinear manifolds beyond the capability of traditional clustering methods. These models project data into a latent space where samples lie in linear subspaces and exploit the self-expressiveness property to uncover intrinsic relationships. Despite their success, existing methods face major limitations: they use shallow autoencoders that ignore clustering errors, emphasize global features while neglecting local structure, fail to model long-range dependencies and positional information, and are rarely applied to 4D spatiotemporal data. To address these issues, we propose A-DATSC (Attention-Guided Deep Adversarial Temporal Subspace Clustering), a model combining a deep subspace clustering generator and a quality-verifying discriminator. The generator, inspired by U-Net, preserves spatial and temporal integrity through stacked TimeDistributed ConvLSTM2D layers, reducing parameters and enhancing generalization. A graph attention transformer based self-expressive network captures local spatial relationships, global dependencies, and both short- and long-range correlations. Experiments on three real-world multivariate spatiotemporal datasets show that A-DATSC achieves substantially superior clustering performance compared to state-of-the-art deep subspace clustering models.