arXiv subjects

Xigang Sun

Publications and source records attributed to Xigang Sun.

3 recordsLinked to original sources

arXiv2025

Huayu: Advanced Real-Time Precipitation Estimation from Geostationary Satellite

As climate change drives increased frequency and intensity of extreme precipitation and flooding worldwide, posing escalating threats to public safety and economic assets, accurate and real-time satellite-based precipitation estimation is essential for operational large-scale hydrometeorological analysis and disaster monitoring. NASA's Integrated Multi-satellitE Retrievals for GPM (IMERG Final Run) combines information from "all" satellite microwave observations with gauge correction and climatological adjustment to produce precipitation estimates at 0.1{\deg} spatial and 30-min temporal resolution. However, its latency of approximately 3.5 months restricts its utility for real-time applications, despite outperforming mainstream satellite precipitation datasets in representing rainfall patterns and variability. We present Huayu, a novel machine learning-based real-time satellite precipitation retrieval system that relies solely on infrared observations from the FengYun-4B geostationary satellite to provide a more accurate precipitation estimate at a finer spatiotemporal resolution (15 min, 0.05{\deg}) over a 120{\deg} by 120{\deg} domain. Performance evaluations demonstrate that Huayu achieves strong consistency with rain gauge observations, yielding a critical success index (CSI) of 0.693 - representing a 3.43% improvement over IMERG Final Run (CSI: 0.670). Experimental results confirm that infrared satellite observations can deliver more accurate precipitation estimates than conventional multi-source algorithms.

physics.ao-ph↗

arXiv2024

Urban Region Pre-training and Prompting: A Graph-based Approach

Urban region representation is crucial for various urban downstream tasks. However, despite the proliferation of methods and their success, acquiring general urban region knowledge and adapting to different tasks remains challenging. Existing work pays limited attention to the fine-grained functional layout semantics in urban regions, limiting their ability to capture transferable knowledge across regions. Further, inadequate handling of the unique features and relationships required for different downstream tasks may also hinder effective task adaptation. In this paper, we propose a $\textbf{G}$raph-based $\textbf{U}$rban $\textbf{R}$egion $\textbf{P}$re-training and $\textbf{P}$rompting framework ($\textbf{GURPP}$) for region representation learning. Specifically, we first construct an urban region graph and develop a subgraph-centric urban region pre-training model to capture the heterogeneous and transferable patterns of entity interactions. This model pre-trains knowledge-rich region embeddings using contrastive learning and multi-view learning methods. To further refine these representations, we design two graph-based prompting methods: a manually-defined prompt to incorporate explicit task knowledge and a task-learnable prompt to discover hidden knowledge, which enhances the adaptability of these embeddings to different tasks. Extensive experiments on various urban region prediction tasks and different cities demonstrate the superior performance of our framework.

cs.AI↗

arXiv2023

Explicit Time Embedding Based Cascade Attention Network for Information Popularity Prediction

Predicting information cascade popularity is a fundamental problem in social networks. Capturing temporal attributes and cascade role information (e.g., cascade graphs and cascade sequences) is necessary for understanding the information cascade. Current methods rarely focus on unifying this information for popularity predictions, which prevents them from effectively modeling the full properties of cascades to achieve satisfactory prediction performances. In this paper, we propose an explicit Time embedding based Cascade Attention Network (TCAN) as a novel popularity prediction architecture for large-scale information networks. TCAN integrates temporal attributes (i.e., periodicity, linearity, and non-linear scaling) into node features via a general time embedding approach (TE), and then employs a cascade graph attention encoder (CGAT) and a cascade sequence attention encoder (CSAT) to fully learn the representation of cascade graphs and cascade sequences. We use two real-world datasets (i.e., Weibo and APS) with tens of thousands of cascade samples to validate our methods. Experimental results show that TCAN obtains mean logarithm squared errors of 2.007 and 1.201 and running times of 1.76 hours and 0.15 hours on both datasets, respectively. Furthermore, TCAN outperforms other representative baselines by 10.4%, 3.8%, and 10.4% in terms of MSLE, MAE, and R-squared on average while maintaining good interpretability.

cs.SI↗