Recent advancements in machine learning have led to a series of breakthroughs in modeling high-dimensional probability densities, from generative modeling, sampling to stochastic optimal control. In this talk, I will discuss how these computational advancements lead to accelerating the sampling of transition paths in molecular systems. In the first part, I will talk about when data is absent, how efficient sampling and optimization can be done with parameterization of the solution by neural networks. The second part of the talk will showcase when data is present, generative approaches can escape the high-cost optimization procedure of complex energy landscapes in transition state search.
Biography:
Yuanqi Du is a PhD student at the Department of Computer Science, Cornell University studying AI and its intersection with Scientific Discovery advised by Prof. Carla P. Gomes. His research interests include Geometric Deep Learning, Probabilistic Machine Learning, Sampling, Optimization, and AI for Science (with a focus on molecular discovery). Aside from his research, he is passionate about education and community building. He leads the organization of a series of events such as the Learning on Graphs conference and AI for Science, Probabilistic Machine Learning workshops at ML conferences and an educational initiative (AI for Science101) to bridge the AI and Science community. https://yuanqidu.github.io/
