Wei Zhong
(Chinese name: é åØ )
I completed my Ph.D. degree of Computer Science at the University of Waterloo in September 2023, advised by Prof. Jimmy Lin.
Started my Ph.D. study at the Rochester Institute of Technology in 2017 Fall, I initially was advised by Prof. Richard Zanibbi, but later paused and transferred to the current program in April 2021 due to the COVID-19 global pandemic.
Before that, I have received my Masterās degree in Electrical and Computer Engineering (advised by Prof. Fang Hui) at the University of Delaware and my B.S. degree from Jiliang University of China in 2015 and 2013, respectively.
In my research, I have been working on creating a math-aware search engine. I have won two best paper awards during my graduate study.
With a broad knowledge in computer science in general, my primary focus is currently in information retrieval, deep learning, and systems.
Interesting Projects
-
Approach Zero: A math-aware search engine combining math expression and regular keyword-based retrieval.
- Ranked #1 by top user votes in years 2020 and 2021 in the Math Stack Exchange Community Promotion Ads.
- The best performance in all math-aware search tasks of CLEF-2022 ARQMath tasks.
- The best unsupervised formula search in the formula retrieval task of ARQMath 2021.
- I have proudly made this project demo online (using Docker swarm) and have successfully kept the site up since 2016 (served more than a million queries).
- TinyNN and MNN: Deep learning projects for my study purpose. I have contributed: Denoising autoencoder (DAE), Restricted Boltzmann Machine with CD-k training, and activation maximization (AM) visualization. I have also documented manual derivations for the gradients/Jacobian matrix of typical layers.
- FlappyBird in VHDL: An interesting course project I made in an embedded system course using VHDL (see Youtube video)
- 8-bit TTLāCPU: Homebrew 8-bit TTL CPU, it successfully runs a for loop to calculate 1 + 2 + ā¦ + 19 in Multisim software simulation.