Atomic Cluster Expansion

Atomic cluster expansion (ACE) is a machine learning technique used to develop accurate and efficient interatomic potentials for materials simulations. Current research focuses on improving ACE model architectures, such as exploring Cartesian coordinate-based expansions and unifying different approaches like ACE and neural equivariant potentials to optimize accuracy and computational efficiency. These advancements enable faster and more precise modeling of material properties, including phonon transport and thermal conductivity, with applications in areas like high-entropy alloys and electronic device design. The resulting potentials are crucial for large-scale atomistic simulations that were previously computationally prohibitive.