About

Hello, I'm Yiqiu (pronounced "yee-chiu"). I work on compilers for machine learning accelerators at Apple, focusing on inference performance and model enablement. My work has spanned new hardware bring-up through production.

I completed my PhD in Computer Science at MIT, advised by Prof. Julian Shun at CSAIL, on parallel algorithms for clustering and computational geometry that are both provably efficient and fast in practice. I earned my MS and BS at Rice University, where I worked with Prof. Anshumali Shrivastava on GPU algorithms for large-scale machine learning.

Selected Publications

• ParGeo: A Library for Parallel Computational Geometry [ESA'22] [arXiv] [Code]

• Fast Parallel Algorithms for Euclidean Minimum Spanning Tree and Hierarchical Spatial Clustering [SIGMOD'21] [arXiv] [Code]

• Theoretically Efficient and Practical Parallel DBSCAN [SIGMOD'20] [arXiv] [Code]

• Randomized Algorithms Accelerated over CPU-GPU for Ultra-High Dimensional Similarity Search [SIGMOD'18] [arXiv] [Code]