Limei Wang

Research Abstract:

With the rapid advancement of artificial intelligence (AI), its applications in scientific research have grown significantly, giving rise to the research area of AI for science (AI4Science). My research focuses on AI for molecular science (molecules, proteins, materials), aiming to build efficient and effective methods to accelerate molecular discovery. Specifically, I work on molecular representation learning using graph neural networks (GNNs). In this talk, I will present my contributions to AI for molecule science from three perspectives: symmetry, expressiveness, and efficiency. First, in order to incorporate 3D structures of molecules, deep learning methods should follow the symmetry of rotation and translation equivariance. In addition, The expressiveness of learned representations is critical for the distinguishing ability of deep learning models, while high efficiency enables fast training and inference, and enhances scalability to large-scale, real-world datasets.

Bio:

Limei Wang a final-year Ph.D. student in the Department of Computer Science & Engineering at Texas A&M University, advised by Prof. Shuiwang Ji. Before TAMU, she received her bachelor’s degree in Automation at University of Science and Technology of China (USTC) in June 2019. Her research interests include machine learning, deep learning, and graph analysis. Specifically, she works on graph/geometric ML for molecules (e.g. molecules, proteins, materials, drug discovery), scalable methods for large-scale graphs (e.g. social network, citation network), and graph generative models. Her work results in publications at top-tier conferences and journals, such as NeurIPS, ICML, ICLR, JMLR, and Bioinformatics, and shows practical values through real-world challenges.