CASC: Causal Adversarial Subspace Clustering for Multivariate Spatiotemporal Data

• URL: http://arxiv.org/abs/2607.21088v1
• Date: Thu, 23 Jul 2026 09:19:17 GMT
• Title: CASC: Causal Adversarial Subspace Clustering for Multivariate Spatiotemporal Data
• Abstract summary: We propose a novel Causal Adrial Subspace Clusteringgeneration (CASC) framework for discovering latent regimes in high-dimensional data.<n>CASC integrates a U-Net-inspired deep adversarial clustering architecture with stacked FAConvLSTM layers to preserve spatial and temporal structure.<n>We also propose two new learning objectives: (1) a Causal Preservation Loss that aligns self-expression with latent causal relationships, and (2) a Dynamic Temporal Subspace Evolution Loss that captures evolving subspace structures and temporal regime transitions in nonstationary environments.
• Score: 0.16206783799607727
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Deep subspace clustering plays a critical role in applications involving multivariate spatiotemporal data, such as sea ice monitoring, disease spread analysis, and tracking neuro-degeneration over time. Despite recent advances, existing methods primarily rely on geometric self-expressiveness, assume static subspace structures, and often fail to capture causal dependencies, local spatial interactions, and long-range temporal dynamics inherent in complex spatiotemporal systems. To address these limitations, we propose a novel Causal Adversarial Subspace Clustering (CASC) framework for discovering evolving latent regimes in high-dimensional spatiotemporal data. CASC integrates a U-Net-inspired deep adversarial clustering architecture with stacked FAConvLSTM layers to preserve spatial and temporal structure while learning robust latent representations. A graph attention transformer-based self-expressive network is introduced to jointly model local spatial relationships, global dependencies, and long-range temporal interactions. Furthermore, we propose two new learning objectives: (1) a Causal Subspace Preservation Loss that aligns self-expression coefficients with latent causal relationships, encouraging clusters to reflect underlying causal processes rather than simple feature similarity, and (2) a Dynamic Temporal Subspace Evolution Loss that captures evolving subspace structures and temporal regime transitions in nonstationary environments. Together, these components transform deep subspace clustering from a correlation-driven paradigm into a causal-temporal regime discovery framework.

Related papers

• Learning Spatio-Temporal Foundation Models from Pure Synthetic Data [85.23980441738811]
  NeoST is a scalable synthetic-training corpus to real-world bias, a latent-space reasoning architecture that generates and iteratively refines multiple future trajectories, and latent-space objectives that emphasize structural dynamics and enable inference-time correction under distribution shifts.<n>Extensive experiments across diverse real-world benchmarks show that NeoST consistently outperforms existing STFMs in diverse real-worldtemporal systems.
  arXiv  Detail & Related papers  (2026-06-27T02:56:46Z)
• M-CaStLe: Uncovering Local Causal Structures in Multivariate Space-Time Gridded Data [0.6054759773528756]
  Causal graph discovery for space-time systems is challenging in high-dimensional gridded data.<n>CaStLe was developed to address that niche under space-time locality and stationarity assumptions.<n>We present M-CaStLe to jointly model local within-identification and cross-variable space-time causal structures in gridded data.
  arXiv  Detail & Related papers  (2026-05-01T04:40:35Z)
• Emergence Transformer: Dynamical Temporal Attention Matters [8.268321002607737]
  The Transformer, a breakthrough architecture in artificial intelligence, owes its success to the attention mechanism.<n>We propose an Emergence Transformer, allowing each component to interact with its own or its neighbors' past states.<n>We show how DTA reshapes social coherence, suggesting strategies to either enhance agreement or preserve plurality.
  arXiv  Detail & Related papers  (2026-04-18T01:10:44Z)
• Unleashing Temporal Capacity of Spiking Neural Networks through Spatiotemporal Separation [67.69345363409835]
  Spiking Neural Networks (SNNs) are considered naturally suited for temporal processing, with membrane potential propagation widely regarded as the core temporal modeling mechanism.<n>We design Non-Stateful (NS) models progressively removing membrane propagation to its stage-wise role. Experiments reveal a counterintuitive phenomenon: moderate removal in shallow layers improves performance, while excessive removal causes collapse.
  arXiv  Detail & Related papers  (2025-12-05T07:05:53Z)
• Attention-Guided Deep Adversarial Temporal Subspace Clustering (A-DATSC) Model for multivariate spatiotemporal data [0.16206783799607727]
  Subspace clustering models are vital for applications such as snowmelt detection sea ice tracking.<n>A graph attention based self-expressive network captures local spatial relationships, global dependencies, and both short- and long-range preserves.<n> Experiments on three real-world datasets show that A-SCSC achieves substantially superior clustering performance compared to state-of-the-art deep subspace clustering models.
  arXiv  Detail & Related papers  (2025-10-20T18:38:26Z)
• RainDiff: End-to-end Precipitation Nowcasting Via Token-wise Attention Diffusion [64.49056527678606]
  We propose a Token-wise Attention integrated into not only the U-Net diffusion model but also the radar-temporal encoder.<n>Unlike prior approaches, our method integrates attention into the architecture without incurring the high resource cost typical of pixel-space diffusion.<n>Our experiments and evaluations demonstrate that the proposed method significantly outperforms state-of-the-art approaches, robustness local fidelity, generalization, and superior in complex precipitation forecasting scenarios.
  arXiv  Detail & Related papers  (2025-10-16T17:59:13Z)
• B-TGAT: A Bi-directional Temporal Graph Attention Transformer for Clustering Multivariate Spatiotemporal Data [0.15763762230817438]
  We present a time-distributed hybrid U-Net autocoder that integrates a Bi-multidimensional Graph Attention Transformer (B-TGAT) to guide efficient temporal clustering.<n>The encoder and decoder are equipped with ConvLSTM2D modules that extract joint spatial-temporal features.<n> experiments show superior cluster separability, temporal stability, and alignment with known climate transitions compared to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-09-16T16:08:21Z)
• Reduced Spatial Dependency for More General Video-level Deepfake Detection [9.51656628987442]
  We propose a novel method called Spatial Dependency Reduction (SDR), which integrates common temporal consistency features from multiple spatially-perturbed clusters.<n>Extensive benchmarks and ablation studies demonstrate the effectiveness and rationale of our approach.
  arXiv  Detail & Related papers  (2025-03-05T08:51:55Z)
• Discovering Latent Causal Graphs from Spatiotemporal Data [23.400027588427964]
  Inferring causal relationships from climate data is a challenging problem with high dimensionality.<n>We present SPACY (tiotemporal CausalY), a novel framework for inference based on latent causal structures.<n> SPACY also identifies key known phenomena from real-world climate data.
  arXiv  Detail & Related papers  (2024-11-08T05:12:16Z)
• A Decoupled Spatio-Temporal Framework for Skeleton-based Action Segmentation [89.86345494602642]
  Existing methods are limited in weak-temporal modeling capability. We propose a Decoupled Scoupled Framework (DeST) to address the issues. DeST significantly outperforms current state-of-the-art methods with less computational complexity.
  arXiv  Detail & Related papers  (2023-12-10T09:11:39Z)
• Anchor-based Multi-view Subspace Clustering with Hierarchical Feature Descent [46.86939432189035]
  We propose Anchor-based Multi-view Subspace Clustering with Hierarchical Feature Descent. Our proposed model consistently outperforms the state-of-the-art techniques.
  arXiv  Detail & Related papers  (2023-10-11T03:29:13Z)
• Supporting Optimal Phase Space Reconstructions Using Neural Network Architecture for Time Series Modeling [68.8204255655161]
  We propose an artificial neural network with a mechanism to implicitly learn the phase spaces properties. Our approach is either as competitive as or better than most state-of-the-art strategies.
  arXiv  Detail & Related papers  (2020-06-19T21:04:47Z)
• A Spatial-Temporal Attentive Network with Spatial Continuity for Trajectory Prediction [74.00750936752418]
  We propose a novel model named spatial-temporal attentive network with spatial continuity (STAN-SC) First, spatial-temporal attention mechanism is presented to explore the most useful and important information. Second, we conduct a joint feature sequence based on the sequence and instant state information to make the generative trajectories keep spatial continuity.
  arXiv  Detail & Related papers  (2020-03-13T04:35:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.