Yuanqing designs graph machine learning methods for drug discovery. During his PhD, he authored a string of machine learning force fields balancing speed and accuracy, including Espalmoa, a model used by pharmaceutical companies to reduce the resource requirement for molecular mechanics force field curation.
A Schmidt Science Fellow, Yuanqing will pivot from Chemistry to Machine Learning to understand the fundamental assumptions and underlying flaws of (equivariant) graph neural networks when applied to molecular tasks. With Professor Mark Tuckerman at the Simons Center for Computational Physical Chemistry and Professor Kyunghung Cho at the Center for Data Science at New York University, Yuanqing is developing an interdisciplinary research program focusing on the inductive biases and first principles of molecular graph machine learning techniques.
Yuanquing hopes his research will boost the hunts’ rationality and efficiency for life-saving therapeutics, which have traditionally relied upon intangible, unquantifiable intuitions of human experts.