Xin Dai

Staff Research Scientist

Research areas: data mining, machine learning, deep learning

Xin Dai, Visa Research scientist.

Biography

Dr. Xin Dai joined Visa Research as a Staff Research Scientist in Feb 2022. Prior to joining Visa, Xin was a Research Assistant at Worcester Polytechnic Institute (WPI), and an AI Intern at Energid-Teradyne. Xin received his Ph.D. in Computer Science from WPI in 2022. His dissertation focuses on designing deep learning models for efficient inference on resources-constrained platforms. He received M.S. and B.S. in Computer Science of Beijing Jiaotong University in 2016 and 2012.

As a member of the Data Analytics team, his research interests are in deep learning and data mining, in particular hard attention models and multi-task learning. Xin has published papers in the top-tier conferences of data mining including the SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD), ACM International Conference on Information & Knowledge Management (CIKM), and SIAM International Conference on Data Mining (SDM).

Publications

  • Xin Dai, Xiangnan Kong, Tian Guo, Xinlu He. FiShNet: Fine-Grained Filter Sharing for Resource-Efficient Multi-Task Learning. In Proc. of CIKM, 2021.
  • Xin Dai, Xiangnan Kong, Tian Guo, Yixian Huang. CiNet: Redesigning Deep Neural Networks for Efficient Mobile-Cloud Collaborative Inference. In Proc. of SDM, 2021.
  • Xin Dai, Xiangnan Kong, Tian Guo, John Boaz Lee, Xinyue Liu, Constance M. Moore. Recurrent Networks for Guided Multi-Attention Classification. In Proc. of KDD, 2020.
  • Xin Dai, Xiangnan Kong, Tian Guo. EPNet: Learning to Exit with Flexible Multi-Branch Network. In Proc. of CIKM, 2020.
  • Xin Dai, Xiangnan Kong, Xinyue Liu, John Boaz Lee, Constance M. Moore. Dual-Attention Recurrent Networks for Affine Registration of Neuroimaging Data. In Proc. of SDM, 2020.