Explore the groundbreaking Crystal Graph Convolutional Neural Networks (CGCNN) algorithm in this 42-minute conference talk by Tian Xie at ICBS2024. Discover how CGCNN, developed in 2018, revolutionized material property prediction based on structural data. Learn about its influence on the evolution of graph neural networks and transformers, and understand its pivotal role in laying the foundation for diffusion-based generative models in materials design. Gain insights into the cutting-edge applications of machine learning in accelerating materials research and development.
Accelerating Materials Design with Crystal Graph Neural Networks
Overview
Syllabus
Tian Xie: Accelerating materials design with crystal graph neural networks #ICBS2024
Taught by
BIMSA
Related Courses
-
Mastering Neural Networks and Model Regularization
-
Neural Diffusion PDEs, Differential Geometry, and Graph Neural Networks
-
Equivariant 3D-Conditional Diffusion Models for Molecular Linker Design
-
Harnessing the Properties of Equivariant Neural Networks for Materials Science
-
Understanding Deep Neural Networks - From Generalization to Interpretability
-
Graph Neural Networks for Multi-Agent Learning
