About me

I am a fifth-year Ph.D. student in the Institute for Data, Systems, and Society (IDSS) at Massachusetts Institute of Technology (MIT), advised by Professor Ali Jadbabaie. I am affilated with the Laboratory for Information and Decision Systems (LIDS). I am broadly interested in dynamical systems, network science, and machine learning on combinatorial structures and relational data. My research leverages graphs to model and reason about dependencies in our interconnected world, aiming to develop principled and practical solutions to complex problems.

On the theory side, I am interested in the understanding attention mechanism in graph neural networks and transformers through a graph-theoretic lens - how it gives rise to the emergence of "intelligence" beyond scaling laws. On the application side, I am interested in developing scalable machine learning solutions for large-scale recommender systems and applying graph-based techniques to uncover patterns and insights in data.

In the summer of 2024, I interned at Snap Research, working on enhancing the scalability of large-scale recommender systems using graph-based approaches under the mentorship of Dr. Tong Zhao.

Prior to coming to MIT, I was an undergraduate student at Washington University in St. Louis. I completed my senior thesis on quantum graphs with Professor Renato Feres.

Please feel free to email me (xinyiwu [at] mit [dot] edu) if you want to chat about research or collaborate.

 

News

2022

      Personal Website: I launch this personal website to record my research history.

Page views:
web counter
Unique visitors:
web counter