Zhibin Wang

Biography

I am a fifth year Ph.D. candidate in Department of computer science and technology, Nanjing University, advised by Prof. Chen Tian and Prof. Sheng Zhong. Now I'm interning at Alibaba DAMO Academy, supervised by Longbin Lai. My research interests mainly include graph computing, mining and learning.

Education

Sept. 2018 - Present Nanjing University

• Ph.D. in Computer Science and Technology

Sept. 2014 - June 2018 Nanjing University of Aeronautics and Astronautics

• B.E. in Computer Science and Technology

Research

• April 2021 - April 2022 Zhibin Wang, Longbin Lai, Yixue Liu, Bing Shui, Chen Tian, Sheng Zhong. I/O-Efficient Butterfly Counting at Scale SIGMOD 23

  • Derive a new I/O lower bound of butterfly counting on hierarchical memory by proposing a new class of algorithms called the semi-witnessing algorithm.
  • Develop IOBufs algorithm that approaches the I/O lower bound.
  • Parallelize IOBufs with a fine-grained approach that carefully trade-off the the I/O-efficiency and parallelism.

• Nov. 2019 - Feb. 2021 Santosh Pandey, Zhibin Wang (co-first author), Sheng Zhong, Chen Tian, Bolong Zheng, Xiaoye Li, Lingda Li, Adolfy Hoisie, Caiwen Ding, Dong Li, Hang Liu. TRUST: Triangle Counting Reloaded on GPUs TPDS

  • Implement a vertex-centric hashing-based triangle counting on GPUs with GPU-friendly hashtable layout.
  • Propose graph reordering to reduce collision in hashtable, virtual combination to balance intra-vertex workload, and degree-aware resources allocation to further reduce collision and balance inter-vertex workload.
  • Scale triangle counting to 1024 GPUs via a graph and workload collaborative partitioning.

• April 2021 - April 2022 Zhibin Wang, Zizhao Zhang, Ziwei Zhang, Shihui Ying, Yue Gao, Yuan Zhang, Sheng Zhong Privacy-preserving Joint Hypergraph Learning Submitted to TPAMI

  • Design two protocols to help multiple parties aggregating their private hypergraph data and conduct joint hypergraph learning in a privacy-preserving manner.
  • Protocols based on light-weight cryptographic primitives, including pseudo-random number generators, sum-preserving matrix confusions, and matrix multiplication-based encryptions.

• June 2022 - Dec. 2022 Longbin Lai, Yufan Yang, Zhibin Wang, Yuxuan Liu, Haotian Ma, Sijie Shen, Bingqing Lyu, Xiaoli Zhou, Wenyuan Yu, Zhengping Qian, Chen Tian, Sheng Zhong, Yeh-Ching Chung, Jingren Zhou GLogS: Interactive Graph Pattern Matching Query At Large Scale Submitted to ATC 23

  • Design a compilation stack that compiles declarative GPM queries into distributed programs.
  • Propose an optimizer that can automatically derive optimal execution plans for GPM queries.
  • Implement a system that allows users to interactively submit and efficiently execute GPM queries at large scale.

Honors & Awards

• 2015 Gold Medal ACM-ICPC Asia Regional Contest(Chang Chun Site)
• 2015 China National Scholarship
• 2016 Outstanding Undergraduate Award CCF(China Computer Federation)