Yinwei Dai

I am a 1st year Computer Science Ph.D. student at Princeton University, working with Prof. Ravi Netravali.

I obtained my M.S.E. and B.S.E. in Computer Science at the University of Michigan and B.S.E in Electrical and Compter Engineering from Shanghai Jiao Tong University

Previously, I worked with Prof. Mosharaf Chowdhury and Prof. Harsha V. Madhyastha on projects related to ML systems and Big data analytics. I was also supervised by Prof. David Fouhey on Computer Vision related projects.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github

My research interests are at the intersection of networked system and machine learning in the context of data-intensive computing, including big data systems for scale-out ML and federated data analytics.

We propose Auxo, a scalable FL system that enable the server to decompose the large-scale FL task into groups with smaller intra-cohort heterogeneity.

We present FedScale, a diverse set of challenging and realistic benchmark datasets to facilitate scalable, comprehensive, and reproducible federated learning (FL) research.

We introduce ModelKeeper, a cluster-scale model service framework to accelerate DNN training, by reducing the computation needed for achieving the same model performance via automated model transformation

Interesting projects I have done.

The project has built a P6-structure pipeline to handle RISC-V instructions. Some advanced features such as superscalar execution are added to enhance processor performance.

Implementation of a full stack compiler including lexical scanning, parsing, abstract syntax trees, semantic analysis, intermediate code generation, MIPS code generation and optimization

The capstone project made by Godzilli Studio

A small puzzel game I made with unity. The player needs to change color space accordingly to pass each level

Reimplementation of classic The Legend of Zelda 1986 with Unity