Changcheng Yuan

I am currently a Computer Science PhD student at Texas A&M University. My research interests sit at the intersection of theory and systems, and I am interested in building high-performance systems grounded in strong theoretical foundations.

I currently work on analog error-correcting codes (Analog ECC) for in-memory analog computing circuits under the supervision of Dr. Anxiao (Andrew) Jiang. My work focuses on theory for Analog ECC design and algorithms for efficient code evaluation and decoding.

I am also interested in parallel computing, especially GPU based algorithms. I completed my M.S. in Computer Science at UC Davis, where I was advised by Dr. John Owens and worked on GPU-based sparse direct solvers and graph partitioning algorithms.

GitHubLinkedInEmail
Portrait of Changcheng Yuan

Research Interests

News

  1. Papers Accepted!

    ISIT今年在广州，好想吃叉烧和烧鹅

Publications

  1. On the Height Profile of Analog Error-Correcting Codes

    Ron M. Roth, Ziyuan Zhu, Changcheng Yuan, Paul H. Siegel, Anxiao Jiang

    IEEE International Symposium on Information Theory (ISIT), 2026 (Accepted. arXiv preprint arXiv:2602.20366.)

    arXivDOI

  2. Fast Sparse Matrix Permutation for Mesh-Based Direct Solvers

    Behrooz Zarebavami, Ahmed H. Mahmoud, Ana Dodik, Changcheng Yuan, Serban D. Porumbescu, John D. Owens, Maryam Mehri Dehnavi, Justin Solomon

    ACM SIGGRAPH, 2026 (Accepted to the conference track. arXiv preprint arXiv:2602.00898.)

    arXiv

Projects

  1. GPU SpMM vs. GEMM Break-even

    Aug. 2025

    A performance study of sparse-dense matrix multiplication against dense GEMM baselines on GPU.

  2. 2D Peridynamics GPU Simulator

    GitHub

    Jun. 2025 - Present

    A GPU-accelerated 2D peridynamics simulator using BLAS-like primitives, tensor operations, and domain decomposition.

  3. Multi-label Graph Mining System

    GitHub

    Sept. 2024 - Mar. 2025

    A graph mining system for comparing heuristics, enumerating patterns, and deduplicating embeddings.

  4. SPGEMM Dataflow Analysis on GPU

    GitHub

    Oct. 2022

    An implementation and analysis of inner-product, outer-product, and row-wise SPGEMM dataflows on GPU.

Internship

  1. Pony.AI

    Software Engineer Intern, Jun. 2019 - Sept. 2019 , Guangzhou, China

    Built a Linux C++/OpenGL dashboard for map and lane rendering, sensor overlays, and interactive controls, and implemented an i18n module with smart language-file loading and unloading to reduce runtime resource usage.