BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.2.3.1//EN TZID:America/New_York X-WR-TIMEZONE:America/New_York BEGIN:VEVENT UID:848@fds.yale.edu DTSTART;TZID=America/New_York:20250327T113000 DTEND;TZID=America/New_York:20250327T130000 DTSTAMP:20250916T142148Z URL:https://fds.yale.edu/events/fds-x-applied-physics-colloquium-grant-rot skoff-stanford-efficient-variational-inference-with-generative-models/ SUMMARY:FDS x Applied Physics Colloquium: Grant Rotskoff (Stanford)\, "Effi cient variational inference with generative models" DESCRIPTION:\nAbstract: Neural networks continue to surprise us with their remarkable capabilities for high-dimensional function approximation. Appl ications of machine learning now pervade essentially every scientific disc ipline\, but predictive models to describe the optimization dynamics\, inf erence properties\, and flexibility of modern neural networks remain limit ed. In this talk\, I will introduce several approaches to both analyzing a nd building generative models to augment Monte Carlo sampling and sampling high-dimensional distributions. I will focus\, in particular\, on two app lications from chemistry: sampling conformational ensembles of disordered protein domains and molecular optimization. I will also introduce a self-d istillation strategy for large scale models that shares conceptually simil arities to preference optimization with \; \;reinforcement learnin g\, but does not require proximal optimization (PPO) and outperforms direc t preference optimization and (DPO). \;\n\n\n\nSpeaker bio: Grant Rots koff studies the nonequilibrium dynamics of living matter with a particula r focus on self-organization from the molecular to the cellular scale. His work involves developing theoretical and computational tools that can pro be and predict the properties of physical systems driven away from equilib rium. Recently\, he has focused on characterizing and designing physically accurate machine learning techniques for biophysical modeling. Prior to h is current position\, Grant was a James S. McDonnell Fellow working at the Courant Institute of Mathematical Sciences at New York University. He com pleted his Ph.D. at the University of California\, Berkeley in the Biophys ics graduate group supported by an NSF Graduate Research Fellowship. His t hesis\, which was advised by Phillip Geissler and Gavin Crooks\, developed theoretical tools for understanding nonequilibrium control of the small\, fluctuating systems\, such as those encountered in molecular biophysics. He also worked on coarsegrained models of the hydrophobic effect and self- assembly. Grant received an S.B. in Mathematics from the University of Chi cago\, where he became interested in biophysics as an undergraduate while working on free energy methods for large-scale molecular dynamics simulati ons.\n\n\n\nhttps://statmech.stanford.edu\n\n\n\nHosted by John Sous.\n CATEGORIES:FDS Events,Colloquium LOCATION:Yale Institute for Foundations of Data Science\, Kline Tower 13th Floor\, Room 1327\, New Haven\, CT\, 06511\, United States X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Kline Tower 13th Floor\, Ro om 1327\, New Haven\, CT\, 06511\, United States;X-APPLE-RADIUS=100;X-TITL E=Yale Institute for Foundations of Data Science:geo:0,0 END:VEVENT BEGIN:VTIMEZONE TZID:America/New_York X-LIC-LOCATION:America/New_York BEGIN:DAYLIGHT DTSTART:20250309T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:EDT END:DAYLIGHT END:VTIMEZONE END:VCALENDAR