|
|
|
About Me
I obtained my Ph.D degree at Department of Computer Science and Engineering in Hong Kong University of Science and Technology (HKUST) in Jan. 2018. Before that, I got my bachelor degree at HuaZhong University of Science and Technology (HUST) in 2013. There, I acted as a software consultant of 'Dian Group' for two years, and was a half-year internship at Multi-media and Communication Lab (McLab) doing about image classification. I am a Google Fellowship (machine learning) winner in 2016.
Research Focus
My research focus on making machine learning faster, more compact and robust.
I wish to develop easy and intuitive methods, which can be used by many others with, perhaps, not much professional knowledge of underneath methods.
- Topics in progress: (a). Automatic machine learning; (b). (Knowledge) graph embedding; (c). Nonconvex optimization for machine learning; and (d). Robust learning with noisy label.
- Previous topics: (a). Faster and more compact convolution methods; (b). Low-rank optimization and modeling; (c). Recommender system; and (d). Network compression.
News
- Open Positions: Intern and full-time opportunities for Machine Learning Research@4Paradigm.
- Our paper on "robust semi-definite programming" is accepted to IJCAI-2019.
- Our paper on "cross-organizational diabetes prediction (with a differential privacy guarantee)" is accepted to IJCAI-2019.
- Our paper "SparseHMM" modeling trajectory sequence is accepted to KDD-2019 (research track).
- Our paper "AutoCross" which proposes AutoML techniques in feature engineering is accepted to KDD-2019 (industrial track).
- Our paper "NORT" (studies the fundamental tensor regularization techniques) is accepted to ICML-2019.
- Our paper "NGMeet" is accepted to CVPR-2019 by oral presentation; try our new model for hyper-spectral image.
- Our paper "NSCaching" is accepted to ICDE-2019; try our new negative sampling method for knowledge graph embedding.