Abstract: This talk wi l give a detailed introduction to design principles for generative models of molecular systems. We wi l first discuss several relevant paradigms for building generative models, including score-based diffusion and continuous normalizing flows. We wi l then detail how both intramolecular and intermolecular constraints inform the architectures appropriate for modeling nanoscale materials. Finaly, we wi l come to the crucial question of how to build scalable training strategies for systems where thermal fluctuations cannot be neglected. The talk will end with a discussion of the most pressing open problems for generative “AI” in the physical sciences.

Reading list:

● https://pubs-acs-org.stanford.idm.oclc.org/doi/ful/10.1021/acs.jpcb.3c0…

● https://pubs-aip-org.stanford.idm.oclc.org/aip/jcp/article/158/12/12412… 0