HKUST DB4AI Group

About Us

DB4AI (Database for Artificial Intelligence) group is dedicated to investigating data management techniques to optimize AI models and systems. Our research is motivated by new technologies and applications. We are particularly interested in the topics of large-scale graph learning, compilation optimization of DNN operators, efficient streaming learning, as well as other theoretical and engineered problems in AI. We are part of the Department of Computer Science and Engineering at the The Hong Kong University of Science and Technology.

Group Members

Faculty

Lei Chen

Professor

Collaborators

Yanyan Shen

Associate Professor, Shanghai Jiao Tong University

Yingxia Shao

Associate Professor, Beijing University of Posts and Telecommunications

Yue Wang

Research Fellow, Shenzhen Institute of Computing Sciences

Graduate Students

Yiming Li

Ph.D Student

Jingzhi Fang

Ph.D Student

Jia Li

Ph.D Student

Xin Zhang

Ph.D Student

Qiqi Zhou

Ph.D Student

Mengxuan Li

Ph.D Student

Shihong Gao

Ph.D Student

Research Topics

Scalable GNN Systems
  • Graph Sampling
  • Distributed Training
  • Knowledge Distillation
Compilation Optimization of DNNs
  • High-Performance Tensor Program Reuse
  • Optimizing DNN Computation Graph via Graph Substitutions
Learning over Data Streams
  • Efficient Continual Learning
  • Scalable Temporal Graph Neural Networks

Publications