Hello! I am a postdoctoral researcher at Foundamental AI Research (FAIR) in Meta, hosted by Lin Xiao. My research interests lie in the intersection of deep learning, optimization and statistics.

I develop theory and methodologies to advance training and inference efficiency and performance for AI models. For training, we designed efficient and robust optimization algorithms (such as BCOS) that achieve comparable performance as Adam but requires less memory and fewer hyper-parameters. For inference, we are developing principled approaches that prune and quantize models during training to achieve maximum inference efficiency without compromising performance.

I recently got my PhD in Operations Research at Cornell University, advised by Damek Davis. Prior to that, I received MMath in Combinatorics and Optimization at University of Waterloo, advised by Steve Vavasis, and BEng in Engineering Systems and Design at Singapore University of Technology and Design.

Outside of office, I enjoy hiking, rock climbing and going out for adventures with my husband Zhi.