 |
Research Interests
|
Biography
I graduated with distinction from the Special Class for the Gifted Young at Xi’an Jiaotong University, China, earning a B.E. degree in Electrical Engineering. As an undergraduate, I visited the University of Wisconsin-Madison and received the Academic Excellence Award. I later conducted research on neuromorphic computing circuits at the University of Maryland. Afterwards, I obtained an MSc in Electrical Engineering with a minor in Computational Neuroscience at EPFL. My master’s research focused on developing theories to characterize stationary points that slow down the training dynamics of neural networks, particularly in non-smooth, non-convex settings. I also discovered human-like visual perceptual patterns and illusions in self-supervised deep learning models. Currently, I am a doctoral student in the Computer Science fellowship program at EPFL, supervised by Prof. Volkan Cevher. My research aims to develop practically motivated theories that enhance the efficiency and performance of large language models.