Abstract: In the era of big data, there has been a surge in availability data containing rich spatial and temporal information, offering valuable insights into dynamic systems processes for applications such as weather forecasting, natural disaster management, intelligent transport systems, precision agriculture. Graph neural networks (GNNs) have emerged powerful tool modeling understanding with dependencies to each other dependencies. There is large amount existing work that focuses on...

Graph Neural Networks (GNNs) have revolutionized molecular property prediction by leveraging graph-based representations, yet their opaque decision-making processes hinder broader adoption in drug discovery. This study introduces the Activity-Cliff-Explanation-Supervised GNN (ACES-GNN) framework, designed to simultaneously improve predictive accuracy and interpretability integrating explanation supervision for activity cliffs (ACs) into training. ACs, defined structurally similar molecules...

Ubiquitous geometric objects can be precisely and efficiently represented as polyhedra. The transformation of a polyhedron into vector, known polyhedra representation learning, is crucial for manipulating these shapes with mathematical statistical tools tasks like classification, clustering, generation. Recent years have witnessed significant strides in this domain, yet most efforts focus on the vertex sequence polyhedron, neglecting complex surface modeling real-world polyhedral objects....

In recent years, graph neural networks (GNNs) and the research on their explainability are experiencing rapid developments achieving significant progress. Many methods proposed to explain predictions of GNNs, focusing "how generate explanations" However, questions like "whether GNN explanations inaccurate", "what if adjust model more accurate have not been well explored. To address above questions, this paper proposes a Explanation Supervision (GNES) <sup...

Spatial interpolation is the task to interpolate targeted index, such as PM2.5 values and temperature, at arbitrary locations based on collected geospatial data. It greatly affects key research topics in geoscience terms of obtaining heterogeneous spatial information (e.g., soil conditions, precipitation rates, wheat yields) for geographic modeling decision-making local, regional, global scales. Point-based data, by ground-level in-situ sensors, serve an important data source this task....

As the societal impact of Deep Neural Networks (DNNs) grows, goals for advancing DNNs become more complex and diverse, ranging from improving a conventional model accuracy metric to infusing advanced human virtues such as fairness, accountability, transparency (FaccT), unbiasedness. Recently, techniques in Explainable Artificial Intelligence (XAI) are attracting considerable attention, have tremendously helped Machine Learning (ML) engineers understanding AI models. However, at same time, we...

Polygon representation learning is essential for diverse applications, encompassing tasks such as shape coding, building pattern classification, and geographic question answering. While recent years have seen considerable advancements in this field, much of the focus has been on single polygons, overlooking intricate inner- inter-polygonal relationships inherent multipolygons. To address gap, our study introduces a comprehensive framework specifically designed representations polygonal...

Multi-source spatial point data prediction is crucial in fields like environmental monitoring and natural resource management, where integrating from various sensors the key to achieving a holistic understanding. Existing models this area often fall short due their domain-specific nature lack strategy for information sources absence of ground truth labels. Key challenges include evaluating quality different modeling relationships among them effectively. Addressing these issues, we introduce...

Spatial prediction is to predict the values of targeted variable, such as PM2.5 and temperature, at arbitrary locations based on collected geospatial data. It greatly affects key research topics in geoscience terms obtaining heterogeneous spatial information (e.g., soil conditions, precipitation rates, wheat yields) for geographic modeling decision-making local, regional, global scales. In situ data, by ground-level sensors, remote sensing satellite or aircraft, are two important data...

Semantic trajectories, which enrich spatial-temporal data with textual information such as trip purposes or location activities, are key for identifying outlier behaviors critical to healthcare, social security, and urban planning. Traditional detection relies on heuristic rules, requires domain knowledge limits its ability identify unseen outliers. Besides, there lacks a comprehensive approach that can jointly consider multi-modal across spatial, temporal, dimensions. Addressing the need...

Spatial autocorrelation and spatial heterogeneity widely exist in data, which make the traditional machine learning model perform badly. domain generalization is a extension of generalization, can generalize to unseen domains continuous 2D space. Specifically, it learns under varying data distributions that generalizes domains. Although tremendous success has been achieved there very few works on generalization. The advancement this area challenged by: 1) Difficulty characterizing...

In the era of big data, there has been a surge in availability data containing rich spatial and temporal information, offering valuable insights into dynamic systems processes for applications such as weather forecasting, natural disaster management, intelligent transport systems, precision agriculture. Graph neural networks (GNNs) have emerged powerful tool modeling understanding with dependencies to each other dependencies. There is large amount existing work that focuses on addressing...

Spatial autocorrelation and spatial heterogeneity widely exist in data, which make the traditional machine learning model perform badly. domain generalization is a extension of generalization, can generalize to unseen domains continuous 2D space. Specifically, it learns under varying data distributions that generalizes domains. Although tremendous success has been achieved there very few works on generalization. The advancement this area challenged by: 1) Difficulty characterizing...