graph neural networks

τ-TCPN: A Causal Topology Self-Organizing Framework via Phase-Coupled Temporal Cut Points — From Deadlock to Emergent Multi-Path Reasoning

We present τ-TCPN (Temporal-Cut-Point Causal Fusion Network), a deterministic, training-free framework for autonomous causal reasoning. Unlike probabilistic neural networks that rely on gradient descent and approximate inference, τ-TCPN models neurons as discrete temporal cut points with hard causal compatibility const...

CHKim-Phys/NuGNN: NuGNN v1.0.0

This contains codes to reproduce graph neural networks for nuclear reactions, or NuGNN.

CHKim-Phys/NuGNN: NuGNN v1.0.0

This contains codes to reproduce graph neural networks for nuclear reactions, or NuGNN.

Spatio-temporal graph neural networks for ground reaction forces prediction from kinematics in individuals with Parkinson’s disease: A comparative study

M-DCATDGN: Enhancing load demand forecasting with an adaptive dynamic graph neural network

Does Graph Structure Earn Its Place in Microservice Root-Cause Analysis? A Controlled Study on RCAEval, and What the Benchmark Was Really Measuring

Graph neural networks dominate recent work on microservice root-cause analysis, yet recent results question whether the graph contributes. Those results compare whole pipelines, so when a flat model wins one cannot tell whether structure is useless or redundant. We run the comparison they imply on RCAEval: three learne...

WidgetGNN: Self-Optimizing HMI Layouts with Graph Neural Networks

GRAPH NEURAL NETWORK (GNN) BASED TOPOLOGY CONTROL IN SELF-ORGANIZING WHEELED ROBOT SWARMS

This study examines a Graph Neural Network (GNN)-based approach for controlling communication topology and coordinated motion in self-organizing wheeled robot swarms under dynamically changing spatial and network conditions. The proposed framework represents robots as graph nodes and wireless communication links as gra...

Pep-PU-GAN: Positive-Unlabeled Adversarial Learning for Peptide Function Prediction

Pep-PU-GAN is introduced, a deep learning framework that combines positive-unlabeled (PU) learning, generative adversarial networks (GANs), and graph neural networks (GNNs) for peptide classification.

Behaviorally prioritized entity-relation structure captures human visual cortical representations of natural scenes

Understanding natural scenes requires identifying visible entities and representing how those entities are related. Recent studies have shown that artificial neural networks (ANNs), large language models (LLMs), and vision language models (VLMs) can predict visual cortical responses to natural images. However, the neur...

Electric Vehicle Charging Demand Forecasting Using Progressive Graph Convolutional Networks With Waterwheel Plant Optimization

Overall, the outcomes prove that PGCN‐WWPA provides more accurate, faster, and scalable EV charging demand forecasting, offering an effective solution for intelligent EM and future SG applications.

τ-TCPN: A Causal Topology Self-Organizing Framework via Phase-Coupled Temporal Cut Points — From Deadlock to Emergent Multi-Path Reasoning

We present τ-TCPN (Temporal-Cut-Point Causal Fusion Network), a deterministic, training-free framework for autonomous causal reasoning. Unlike probabilistic neural networks that rely on gradient descent and approximate inference, τ-TCPN models neurons as discrete temporal cut points with hard causal compatibility const...