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
Thursday, September 8, 2022, 15:00–17:30, H1
Machine learning methods have gained a prominent spot in the research of materials and molecules, especially in the context of the atomic-scale modeling of their properties. The growing understanding of how machine-learning methods should be adapted to the specific requirements of the field is making them progressively more effective and easy to use. Machine-learning techniques use the predictions of electronic-structure theory to train surrogate models that can compute the same properties with similar accuracy at much reduced cost. The combination of physics-based and data-driven paradigms is extending dramatically the reach of electronic-structure theory, as its predictive accuracy can now be applied to more complex, larger-scale problems and longer timescales. The field has been evolving so fast that in the past years we have witnessed considerable breakthroughs enabled by this combination: first-principles accuracy assessment of finite-temperature thermodynamics, including also subtle effects such as quantum nuclear fluctuations, has become commonplace; predictions of microscopic quantities beyond the interatomic potential energy are making it possible to incorporate functional properties into a fully-predictive machine-learning framework; inverse design and generative models are simplifying the search of configurational and composition spaces for compounds with optimal performance; including information and training data calculated from methods that go beyond density functional theory allows to make predictions systematically improvable.
![]() |
15:00 | SYES 1.1 | Invited Talk: Machine-learning-driven advances in modelling inorganic materials — •Volker L. Deringer |
![]() |
15:30 | SYES 1.2 | Invited Talk: Machine-Learning Discovery of Descriptors for Square-Net Topological Semimetals — •Eun-Ah Kim |
![]() |
16:00 | SYES 1.3 | Invited Talk: Four Generations of Neural Network Potentials — •Jörg Behler |
![]() |
16:30 | SYES 1.4 | Invited Talk: Using machine learning to find density functionals — •Kieron Burke |
![]() |
17:00 | SYES 1.5 | Invited Talk: Coarse graining for classical and quantum systems — •Cecilia Clementi |