Regensburg 2022 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

SYES: Frontiers of Electronic-Structure Theory: Focus on Artificial Intelligence Applied to Real Materials

SYES 1: Frontiers of Electronic-Structure Theory: Focus on Artificial Intelligence Applied to Real Materials

SYES 1.4: Invited Talk

Thursday, September 8, 2022, 16:30–17:00, H1

Using machine learning to find density functionals — •Kieron Burke — University of California, Irvine, USA

Over the past decade, advances in machine learning have led to the creation of new approximate density functionals. I will review this area, with an emphasis on very recent developments. How do such functionals compare to those of human design? What are their advantages and their limitations? For example, can they work for strongly correlated systems? I will consider both the exchange-correlation energy used in Kohn-Sham DFT and the non-interacting kinetic energy functional, needed to bypass the KS equations.

• How Well Does Kohn-Sham Regularizer Work for Weakly Correlated Systems? B. Kalita, R. Pederson, J. Chen, L. Li, and K. Burke, J. Phys. Chem. Lett (2022).

• Machine learning and density functional theory R. Pederson, B. Kalita, and K. Burke, Nat. Rev. Phys. (2022).

• Kohn-Sham Equations as Regularizer: Building Prior Knowledge into Machine-Learned Physics L. Li, S. Hoyer, R. Pederson, R. Sun, E. Cubuk, P. Riley, and K. Burke, Phys. Rev. Lett. 126, 036401 (2021).

• Using Machine Learning to Find New Density Functionals B. Kalita and K. Burke, Article in Roadmap on Machine Learning in Electronic Structure (2022).