Seminar Series - Exploratory Research at Google Accelerated Science. Dr Li Li, Google.

Tuesday, April 28 at 2 p.m. Webinar (zoom link will be posted soon)

Abstract: In this talk, I will describe three exploratory research projects I have been working on at Google Accelerated Science team (https://research.google/teams/applied-science/gas/). Our mission is to produce breakthroughs in natural sciences by applying Google technologies, including machine learning, iterative prediction/experimentation in large combinatorial spaces, and large scale analysis and computation. The first project is neural-guided symbolic regression that can efficiently generate expressions from data by incorporating prior knowledge such as asymptotic constraints. The second project is quantum optimization with a novel Gibbs objective function and ansatz architecture search. The last project presents a framework to optimize molecules by combining domain knowledge of chemistry and deep reinforcement learning techniques.