Education & Work

Zeyuan on the photo day of MSR AI
@ Microsoft Research, 2017

Tsinghua University

Personal Information

Research Areas

Deep Learning, Machine Learning, Optimization, Algorithms

Research Interests

In recent years, I work on the mathematical foundations of Deep Learning, where the ultimate goal is to turn black magic into scientific theorems, and then design more principled algorithms for a better, safer, more economical world of artificial intelligence. In my past life, I also worked on the theory of machine learning and optimization, and apply them to theoretical computer science, operations research, and statistics. I also am broadly interested in the mathematical modeling for physical, social, economic, and biological systems.

As part of my research, I also maintain my own PyTorch code base for training ResNet, DenseNet, Transformers, BERT, GPT-2, Vision Transformers, and apply them to vision tasks, Q&A tasks such as SQUAD, language modeling tasks such as WikiText103, text generation tasks such as E2E, WikiSQL, etc.





Some Awards

A family photo of my gold and silver medals
A family photo of my gold and silver medals

In algorithm competitions, I was fortunate to win a few awards in my past life, including two IOI gold medals, a USACO world champion, an ACM/ICPC world-final gold medal, a Google Codejam world runner-up, and a USA MCM Top Prize.

In research, I used to be supported by a Microsoft Young Fellow Award, a Simons Student Award and a Microsoft Azure Research Award.

For a full list, click here.