graph neural networks

Robust and Privacy-Preserving Decentralized Online Federated Learning for Streaming Data With Outliers

This article addresses the challenging problem of online federated learning (FL) over streaming data in a decentralized communication network. To enable rapid adaptation to new observations, we develop novel nonparametric model-based local training, not deep neural network (DNN)-based approaches as adopted in most prev...

Snapshots of Graph Neural Networks in Asset and Investment Management: An Overview for Practitioners

Multi-station tidal level forecasting based on a novel spatio-temporal graph convolutional neural network

Reconstruction of Shower-like Events in NEON Using Likelihood and Graph Neural Network Methods

The Neutrino Observatory in the Nanhai (NEON) is a proposed deep-sea neutrino telescope deployed in the South China Sea. Accurate reconstruction of shower-like events is crucial for neutrino energy measurements and multi-messenger astronomy, yet it poses significant challenges due to seawater optical attenuation, irreg...

Contextual Graph Neural Networks with Dynamic Attention Mechanisms

Graph Neural Networks (GNNs) have demonstrated remarkable success in various domains, including social network analysis, drug discovery, and recommendation systems. However, a significant limitation of many existing GNN models lies in their reliance on static attention mechanisms. These mechanisms often assign fixed we...

Multi-Scale Spatio-Temporal Graph Neural Networks for Urban Traffic Flow Prediction

Accurate urban traffic flow prediction is crucial for intelligent transportation systems, urban planning, and resource allocation. Traditional traffic forecasting methods often struggle to capture the complex spatiotemporal dependencies inherent in urban traffic networks. This paper proposes a novel approach utilizing...

##基于图神经网络的药物分子设计

This paper explores the application of Graph Neural Networks (GNNs) for automated drug molecule design. Traditional drug discovery relies heavily on empirical methods and serendipitous findings, often suffering from high costs and low success rates. We propose a novel approach that leverages the inherent structural inf...

Graph Neural Networks for Hardware Design Space Exploration

This paper investigates the application of Graph Neural Networks (GNNs) to the domain of hardware design space exploration. Traditional design exploration methods often struggle to effectively capture and optimize the intricate dependencies inherent in complex hardware architectures. We propose a novel framework utiliz...

Graph Representation Learning for Protein-Protein Interaction Networks with Temporal Dynamics

Predicting protein function and understanding protein-protein interactions (PPIs) are central challenges in modern biology. Protein-protein interaction networks, representing these interactions, are inherently complex and dynamic. Traditional methods often struggle to capture the temporal evolution of these networks, f...

Relational Reinforcement Learning with Graph Neural Networks for Robotics

This paper presents a novel approach to robot control in complex, multi-agent environments by leveraging relational reinforcement learning (RRL) and graph neural networks (GNNs). The core idea is that robots benefit significantly from understanding and reasoning about the relationships between themselves and their surr...

Neural Time Flow Mapping

This paper proposes a novel approach to analyzing neural activity termed "Neural Time Flow Mapping." The core idea is to leverage the temporal information inherent in neuronal activity patterns to establish dynamic time flow maps between neurons. This allows for predictive modeling and a deeper understanding of neural...

Multi-Scale Spatio-Temporal Graph Neural Networks for Urban Traffic Flow Prediction

Accurate urban traffic flow prediction is crucial for intelligent transportation systems, urban planning, and resource allocation. Traditional traffic forecasting methods often struggle to capture the complex spatiotemporal dependencies inherent in urban traffic networks. This paper proposes a novel approach utilizing...