Email: <1st name>ster at gmail
I am currently a software engineer at Meta.
Previously I have worked at JPMorgan Chase, Yahoo! Labs, Google, JD.com Silicon Valley Research Center, Moveworks, and Coupang.
I graduated from Stanford Mathematics department in 2012, under the advisory of Prof Persi Diaconis.
Before that I spent one year at University of Utrecht in the Netherlands studying symplectic geometry and mathematical physics.
I graduated from University of Georgia with a B. A. in pure math in 2007.
My research interests lie mainly in the following
Markov chain convergence, especially related to finite and simple Lie groups
Locality Sensitive Hashing
Machine Learning applications in Search Ranking and Retrieval
Matrix theory and linear algebra
Thibaut Horel: Yahoo! Labs Summer 2014
Jingtian Gu: JD.com Silicon Valley Research Center Summer 2020
Haocheng Han: Moveworks Summer 2022
Conference Reviewer Roles
DLP-KDD 2020
AAAI 2021
WebConf 2022
Total variation bound for Kac's random walk
PhD thesis: MIXING TIME OF MARKOV CHAINS ON FINITE AND COMPACT LIE GROUPS
Kac's random walk on the special orthogonal group mixes in polynomial time
Unitary matrix digraphs and minimum semidefinite rank (jointly with Lon Mitchel, Sivaram Narayan)
Maximally consistent sampling and the Jaccard index of probability distributions (jointly with Ryan Moulton)
Cut-off phenomenon in the uniform plane Kac walk (jointly with Bob Hough)
From Semantic Retrieval to Pairwise Ranking: Applying Deep Learning in E-commerce Search (jointly with Rui Li, et al)
A unified neural network approach to e-commerce relevance learning (jointly with Yue Shang, et al)
Towards Personalized and Semantic Retrieval: An End-to-EndSolution for E-commerce Search via Embedding Learning (jointly with Han Zhang, Wenyun Yang, et al)
Cost Minimization for Heterogeneous Systems with Gaussian Distribution Execution Time (jointly with Meikang Qiu, Wenyun Dai)
On the Number of Turns in Reduced Random Lattice Paths (jointly with Weijun Xu)
Classification of commutative zero-divisor semigroup graphs (jointly with Lisa DeMeyer, et al)
Smallest Gaps Between Eigenvalues of Random Matrices With Complex Ginibre, Wishart and Universal Unitary Ensembles (jointly with Dai Shi)
Parallel News-Article Traffic Forecasting with ADMM (jointly with Stratis Ioannidis, et al)
Adversarial Mixture Of Experts with Category Hierarchy Soft Constraint (jointly with Zhuojian Xiao, Guoyu Tang, et al)
Recipient of 5+ peer bonuses at Google Inc
JD Search and Recommendation Best Mentor Award (最佳训导师）of 2019
JD Search and Recommendation Most Positive Value Award (价值观奖）of 2020
JD Search and Recommendation Craftsmanship Award (工匠精神奖) of Oct 2021
Implemented Consumer/Producer queue for IO bound neural training in C++
Implemented Product Quantization Decoder in Java
Ported FAISS Library in Java using JNI
Built various Machine Learning data/training/inference pipelines, in tensorflow 1.x and pytorch
Implemented Near Zero Hash Collision with Torchrec Sharded Embedding Tables
Mixture-of-Depth
Column-wise to Row-wise sharding conversion in Torchrec