王定宇 (Dingyu Wang)

About Me

I'm a PhD student in theoretical computer science at University of Michigan, graduating next year. My advisor is Seth Pettie. I'm currently looking for jobs next year (2025).

Contact me by email: wangdy[at]umich[dot]edu

My research mainly consists of analysis and construction of streaming algorithms. Streaming algorithms are, in short, very compact data structures for monitoring large and usually distributed data streams. In addition to theoretical guarantees, I also care a lot about their simplicity and practical applications.

Manuscripts/In Submission

Multi-dimensional Approximate Counting, Dingyu Wang

Universal Perfect Samplers for Incremental Streams, Seth Pettie and Dingyu Wang

Universal Harmonic Sketching over Turnstile Streams, Dingyu Wang

Probabilistic Counting in Generalized Turnstile Models, Dingyu Wang

Publications

Better Cardinality Estimators for HyperLogLog, PCSA, and Beyond (PODS23), Dingyu Wang and Seth Pettie

Non-Mergeable Sketching for Cardinality Estimation (ICALP21), Seth Pettie, Dingyu Wang and Longhui Yin

Information Theoretic Limits of Cardinality Estimation: Fisher meets Shannon (STOC21), Seth Pettie and Dingyu Wang

Optimal Protocols for 2-Party Contention Resolution (SSS21), Dingyu Wang

Fun Stuff

• A small platform game: Moment Master