graph neural networks

A Physics-First Hamiltonian Latent Model for Twelve Universality Classes of Fluid Behaviour

FEM-ANN-GNN modeling of MHD Darcy–Forchheimer Boger nanofluid flow with coupled heat–mass transfer in a Y-shaped hourglass cavity

This study aims to numerically investigate steady two-dimensional MHD Darcy–Forchheimer mixed convection flow of a Boger nanofluid (Cu/H2O) with coupled heat and mass transfer inside a Y-shaped hourglass cavity containing a cylindrical obstacle, incorporating thermal radiation, Cattaneo–Christov heat flux, Soret ef...

CoTAR: Topology and Atomic State Reconstruction in Condensed Phases

Abstract Universal machine learning interatomic potentials (uMLIPs) enable performing condensed-phase molecular dynamics (MD) simulations with accuracy approaching that of first-principles; however, their lack of explicit molecular topology limits bond-aware analysis and reconnection to classical force fields. This stu...

A Graph-based Neural-network Surrogate Model for an Accelerating Semianalytical Model of Galaxy Formation and Evolution

Abstract Understanding how galaxy populations emerge and evolve from the growth of dark-matter structures is a central challenge in galaxy formation theory. Semianalytic models (SAMs) provide an efficient framework to address this problem, but exploring large ensembles of merger trees across broad parameter spaces rema...

Tree-like pairwise interaction networks

Abstract Modeling feature interactions in tabular data remains a key challenge in predictive modeling, for example, as used for insurance pricing. This paper proposes the tree-like pairwise interaction network (PIN), a novel neural network architecture that explicitly captures pairwise feature interactions through a sh...

AI for Smart Healthcare

Healthcare is a domain where AI, and specifically, deep learning, has shown enormous promise. However, this domain also presents unique challenges in terms of safety, interpretability, and clinical integration. In this chapter, we review core neural architectures-convolutional networks, recurrent networks, graph neural...

REIGN: Regime-Enhanced Intelligent Granger Network for Nonstationary Causal Discovery

The Regime-Enhanced Intelligent Granger Network is proposed, a novel causal discovery framework specifically engineered for nonstationary time series and achieves an Optimal F1 score of 0.529, substantially outperforming PCMCI+ and PC in structural edge recovery.

Identifier-Derived Signal and Sanitization Degeneracy in Graph Learning: A Sufficiency-Aware Audit

# Suggested Zenodo metadata Use these fields when creating the Zenodo record. Replace placeholders as needed. **Title** Reproducibility package for *Identifier-Derived Signal and Sanitization Degeneracy in Graph Learning: A Sufficiency-Aware Audit* **Creator** Mücahit Soylu — Inonu University, Faculty of Engineering, D...

Structural Representation in Crystal Property Prediction: A Controlled Benchmark on Band Gaps

This Zenodo record contains the complete reproducibility package for the study Structural Representation in Crystal Property Prediction: A Controlled Benchmark on Band Gaps. The benchmark evaluates how structural representation, model capacity, and training-data availability affect crystal band-gap prediction under a c...

Research data for "Online Decentralized Estimation of Network Connectivity in Cooperative Robotic Systems Using Graph Neural Networks"

This repository contains the datasets, analysis scripts, and experiment configuration used to generate the results presented in our research paper "Online Decentralized Estimation of Network Connectivity in Cooperative Robotic Systems Using Graph Neural Networks". This is a condensed release containing only the materia...

Molecular Solubility, Hydration Shells, and Inverse Spectral Tomography via 3D Lone-Pair Fiedler Vector Optimization

Predicting aqueous solubility (log S), co-crystallization, and solvation dynamics remains one of the formidable bottlenecks in drug discovery, formulation engineering, and physical chemistry. Conventional approaches rely either on computationally expensive quantum mechanical methods (DFT/MD) or black-box machine learni...

Artificial Intelligence-Driven Design and Optimization of Advanced Drug Delivery Systems: From Formulation Prediction to Clinical Translation: A Comprehensive Review

Advanced drug delivery systems (DDS) — including polymeric and lipid nanoparticles, lipid nanoparticles (LNPs) for nucleic acid therapeutics, solid oral dosage forms, and three-dimensional (3D)-printed and stimuli-responsive platforms — require the simultaneous optimization of dozens of interacting formulation and proc...