graph neural networks

Addressing spatial indistinguishability in spatiotemporal prediction via optimal transport-guided masking

STOT is a self-supervised framework that resolves spatiotemporal ambiguity via structured masking guided by optimal transport and a similarity-aware metric for dynamic inter-node relationships and an optimal transport-based masking strategy to emphasize ambiguous positions during pre-training.

GR-MAPPO: A Graph-Enhanced Reinforcement Learning Framework for Narrow-Beam Directional Neighbor Discovery in UAV Networks

Directional communication has emerged as a key enabling technology for next-generation uncrewed aerial vehicle (UAV) networks to achieve extended transmission ranges and high spectral efficiency. However, the consequent reliance on extremely narrow beamwidths imposes stringent spatial constraints that result in highly...

Memory-Aware Joint Optimisation of Partitioning and Scheduling for Pipeline-Parallel Training

OptPipe is presented, a unified framework that jointly optimises partitioning and scheduling for pipeline parallelism and introduces a memory-aware directed acyclic graph (DAG) that captures both task dependencies and the lifetime of intermediate tensors, enabling explicit reasoning about the trade-off between executio...

Graph Convolutional Networks Combining Dynamic Aggregation of Adjacency Information for Traffic Flow Prediction

Accurate traffic flow prediction is crucial for intelligent transportation systems (ITS), especially in traffic management and route planning. Although spatiotemporal graph convolutional networks are widely used for traffic flow prediction, the simple network graph structure is not sufficient to extract periodic tempor...

GraphGS: Mitigating Pseudo-Label Shift in Imbalanced Node Classification via Balanced Feature Propagation and Minority Node Selection

Graph-structured data in real-world applications often grapple with class imbalance, where underrepresented minority nodes result in biased predictions and diminished learning performance. Although pseudo-labeling methods offer a promising solution for addressing class imbalance, they remain susceptible to pseudo-label...

A Physics-Informed Topology-Adaptive Graph Convolutional Network for Harmonic Source Location in Three-Phase Distribution Networks With Renewable Energy Integration

The large-scale integration of renewable energy sources has led to a significant increase in the number of harmonic sources within distribution networks. Concurrently, the altered supply modes introduced by renewable integration have caused dynamic changes in the network topology. Therefore, this paper proposes a three...

HGT-PPO: A Hybrid Graph-Transformer Approach for Large-Scale DAG Task Scheduling in SAGIN

Space–air–ground integrated network (SAGIN) provides a promising computing infrastructure for 6G applications, but scheduling large-scale directed acyclic graph (DAG) tasks in such networks remains challenging due to dynamic topology, heterogeneous resources, and complex intertask dependencies. This article investigate...

NeuroSchedule2.0: A Novel GNN-Based Scheduling Method With RL-Based Preprocessing Optimization for High-Level Synthesis

During high-level synthesis (HLS), scheduling is a critical step that determines the execution order of operations and directly affects circuit performance. However, existing scheduling methods suffer from either poor solution quality or limited scalability for complex designs containing hundreds of operations. This ar...

HSAGNN: A hybrid structure and attribute graph neural network with attention fusion for link prediction

A DUAL-BRANCH NEURAL FORECASTING ARCHITECTURE WITH SESSION BEHAVIORAL PROFILING AND ELASTIC WEIGHT CONSOLIDATION FOR ADAPTIVE CLOUD RESOURCE ALLOCATION

The AI-driven Adaptive Resource Allocation Engine (AARAE) provides proactive resource control for multi-tenant cloud environments. Reactive autoscaling systems suffer from allocation lag: scaling actions trigger only after hypervisor metrics cross predefined thresholds. AARAE extracts early-warning behavioral signature...

Artificial Intelligence in CPU and ASIC Design: Applications, Design-Space Exploration, and Engineering Constraints

Modern CPU and application-specific integrated circuit (ASIC) development requires engineers to search large design spaces while satisfying power, performance, area, timing, and correctness constraints. Artificial intelligence (AI) is increasingly being integrated into electronic design automation (EDA) to accelerate p...

Feature Augmented SuperHyperGraph Neural Networks for Hierarchical Epidemic Modeling, Environmental Perturbations, and Source Identifiability

Abstract: Infectious disease transmission involves interactions occurring at multiple organizational levels, including individuals,households, communities, and regions. Conventional graph and hypergraph neural networks represent pairwise and higher order interactions, but they do not directly provide a recursive repres...