Symmetries in Machine Learning for Materials Science

Explore the intersection of machine learning and materials science in this one-hour lecture by Ryan Adams from Princeton University. Delve into the importance of periodic structures in materials and meta-materials, focusing on crystalline solids and cellular-solid meta-materials. Learn about innovative techniques for incorporating crystallographic group invariance into machine learning models, both in linear and nonlinear representations. Discover how these symmetry-aware models can enhance the solution of the Schrödinger equation for crystalline solids, improve generative modeling of materials, and expand design spaces for cellular mechanical meta-materials. Gain insights into cutting-edge approaches that strengthen the bond between artificial intelligence and materials science, potentially leading to new material discoveries and advancements in the field.

Symmetries in Machine Learning for Materials Science