InterCat talk

 

Speaker: Julia Westermayr, University of Leipzig, Germany

Title: Machine learning for chemical discovery
 

Abstract:

High-throughput screening of reaction conditions and electronic properties of molecules plays a crucial role in chemical industry. However, the high combinatorial complexity of the various parameters affecting molecular properties leaves unguided searches in chemical space highly inefficient. In this talk, we will discuss various ways of how machine learning can facilitate and advance chemical discovery. Therefore, we will introduce predictive models for electronic properties of molecules and materials [1,2] and discuss their integration with generative learning [3] and reinforcement learning [4] for advanced molecular design and reaction optimization, respectively.

References: 

[1] Julia Westermayr, Michael Gastegger, Dóra Vörös, Lisa Panzenboeck, Florian Joerg, Leticia González, and Philipp Marquetand, Nat. Chem. 2022, 14(8), 914-919. 

[2] Julia Westermayr, Reinhard J. Maurer,Chemical Science 2021 12 (32), 10755-10764

[3] Julia Westermayr, Joe Gilkes, Rhyan Barrett,Nat. Comput. Sci. 2023, 3(2), 139-148. 

[4] Rhyan Barrett and Julia Westermayr, J. Phys. Chem. Lett. 2024, 15, 349-356