graph neural networks

Kinematic Neural Networks: A New Paradigm for Embodied Mechanical Intelligence

Physical Artificial Intelligence and Mechanical Neural Networks (MNNs) embed information processing in physical structures. This paper investigates a complementary route based on deterministic rigid-body kinematics. We formulate a Kinematic Neural Network (KNN) as a layered linkage computation graph whose geometric par...

AI-Driven Logic Synthesis in Electronic Design Automation: A Review of Its Implications for Computer Architecture and Organization

This paper reviews how artificial intelligence is being applied to logic synthesis in Electronic Design Automation (EDA) and what this means for Computer Architecture and Organization (CAO). Using the work of Astillero on AI-driven logic gate synthesis as its central source, it compares conventional rule-based synthesi...

Deep learning assisted screening of light-responsive piezoelectric crystals for next-generation photoacoustic and photothermal spectroscopy

Introduction High piezoelectric materials are promising alternative resonator materials for quartz-enhanced photoacoustic spectroscopy (QEPAS) and laser-induced thermoelastic spectroscopy (LITES), as enhanced piezoelectric response directly improves acoustic detection sensitivity. However, the discovery of materials co...

Micromagnetic_GNN

Code and data for training graph neural networks on micromagnetic grain structures to predict coercivity and maximum energy product (BH_max) of hard magnetic microstructures.

Supplementary Information for "Prediction of Solvent Aggregation in CO2-Loaded Aqueous Amines Using Machine Learning Molecular Dynamics Simulations"

Computational details for training data generation, graph neural network-based interatomic potentials (GNN-IP) development and validation. The file also includes system compositions, aggregate spatial distributions, pair distribution functions, aggregation energy estimation, and mean squared displacements for CO2-loade...

Neural Networks and Deep Learning

Neural networks have moved from relatively simple computational models to a central technology for modern artificial intelligence. This paper presents a structured review of how neural-network methods have developed, how different architectures address different forms of data, and where deep learning is being applied i...

Recurrent neural network-based collaborative perception method for robot multimodal sensors

With the advancement of industrial automation and intelligent manufacturing, robots are increasingly requiredtoperform collaborative perception of their own operating status and external working environment in scenarios suchasassembly, inspection, operation & maintenance and human-robot collaboration. However, multi-mo...

Transferable Atom-Pair Graph Neural Network for Few-Shot Reconstruction of Rigid Homobilayer Sliding Potential-Energy Surfaces

Abstract Interlayer potential-energy surfaces (PESs) govern stacking stability, sliding barriers, and tribology in two-dimensional bilayers, but first-principles mapping is expensive. We introduce an atom-pair graph neural network (APGNN) whose nodes are interlayer atom pairs, placing the representation directly on the...

Research on multirobot SLAM path planning based on deep learning

Multi-robot cooperative SLAM is a core technology for autonomous exploration in large-scale unknown environments. However, the information isolation between traditional SLAM and path planning modules leads to poor real-time performance and low collaborative efficiency. This paper proposes an end-to-end multi-robot SLAM...

SympGal-NODE: Continuous-Time Symplectic Neural ODEs for Relativistic Stellar Swarms: Predicting Escape Times, Exit Trajectories, and Tidal Horizons

Towards robust Graph Neural Networks: A comprehensive strategy against graph and label noise

The increasing use of graph data in emerging domains has highlighted the importance of Graph Neural Networks (GNNs). However, traditional GNNs are highly susceptible to both graph and label noise, leading to degraded and unstable performance under noisy conditions. To address this challenge, we propose a R obust S truc...

Prediction of Solvent Aggregation in CO2-Loaded Aqueous Amines Using Machine Learning Molecular Dynamics Simulations

Aggregation of ion pairs formed during CO2 capture may govern absorption thermodynamics and kinetics, but it is often overestimated by fixed-charge classical force fields. Here, we demonstrate the promise of graph neural network interatomic potentials, which offer quantum-level accuracy, for directly evaluating the agg...