STRAP: Spatio-Temporal Pattern Retrieval for Out-of-Distribution Generalization

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Haoyu, Zhang, Wentao, Miao, Hao, Jiang, Xinke, Fang, Yuchen, Zhang, Yifan
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917004014780416
author Zhang, Haoyu
Zhang, Wentao
Miao, Hao
Jiang, Xinke
Fang, Yuchen
Zhang, Yifan
author_facet Zhang, Haoyu
Zhang, Wentao
Miao, Hao
Jiang, Xinke
Fang, Yuchen
Zhang, Yifan
contents Spatio-Temporal Graph Neural Networks (STGNNs) have emerged as a powerful tool for modeling dynamic graph-structured data across diverse domains. However, they often fail to generalize in Spatio-Temporal Out-of-Distribution (STOOD) scenarios, where both temporal dynamics and spatial structures evolve beyond the training distribution. To address this problem, we propose an innovative Spatio-Temporal Retrieval-Augmented Pattern Learning framework,STRAP, which enhances model generalization by integrating retrieval-augmented learning into the STGNN continue learning pipeline. The core of STRAP is a compact and expressive pattern library that stores representative spatio-temporal patterns enriched with historical, structural, and semantic information, which is obtained and optimized during the training phase. During inference, STRAP retrieves relevant patterns from this library based on similarity to the current input and injects them into the model via a plug-and-play prompting mechanism. This not only strengthens spatio-temporal representations but also mitigates catastrophic forgetting. Moreover, STRAP introduces a knowledge-balancing objective to harmonize new information with retrieved knowledge. Extensive experiments across multiple real-world streaming graph datasets show that STRAP consistently outperforms state-of-the-art STGNN baselines on STOOD tasks, demonstrating its robustness, adaptability, and strong generalization capability without task-specific fine-tuning.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19547
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle STRAP: Spatio-Temporal Pattern Retrieval for Out-of-Distribution Generalization
Zhang, Haoyu
Zhang, Wentao
Miao, Hao
Jiang, Xinke
Fang, Yuchen
Zhang, Yifan
Machine Learning
Artificial Intelligence
Spatio-Temporal Graph Neural Networks (STGNNs) have emerged as a powerful tool for modeling dynamic graph-structured data across diverse domains. However, they often fail to generalize in Spatio-Temporal Out-of-Distribution (STOOD) scenarios, where both temporal dynamics and spatial structures evolve beyond the training distribution. To address this problem, we propose an innovative Spatio-Temporal Retrieval-Augmented Pattern Learning framework,STRAP, which enhances model generalization by integrating retrieval-augmented learning into the STGNN continue learning pipeline. The core of STRAP is a compact and expressive pattern library that stores representative spatio-temporal patterns enriched with historical, structural, and semantic information, which is obtained and optimized during the training phase. During inference, STRAP retrieves relevant patterns from this library based on similarity to the current input and injects them into the model via a plug-and-play prompting mechanism. This not only strengthens spatio-temporal representations but also mitigates catastrophic forgetting. Moreover, STRAP introduces a knowledge-balancing objective to harmonize new information with retrieved knowledge. Extensive experiments across multiple real-world streaming graph datasets show that STRAP consistently outperforms state-of-the-art STGNN baselines on STOOD tasks, demonstrating its robustness, adaptability, and strong generalization capability without task-specific fine-tuning.
title STRAP: Spatio-Temporal Pattern Retrieval for Out-of-Distribution Generalization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2505.19547