CV

Education

• B.S. in Aerospace Engineering, Xi’an Jiao Tong University, 2010
• M.S. in Aerospace Engineering, Shanghai Jiao Tong University, 2013
• Ph.D in Applied and Computational Mathematics, Florida State University, 2021 (expected)

Work experience

• May 2020 - Aug 2020: Machine Learning Research Intern, Kwai.inc
  • Ads ranking on video-sharing app with 200 million DAU
  • E-commerce ads content understanding (image Scene Text Understanding and multimodal learning)
• May 2019 - Aug 2019: SDE intern, Weride.ai
  • Designed an intelligent human driver agent which has multiple behavioral intentions
  • Prototyped the application of reinforcement learning algorithm (DP, DQN) to solve lane change problem for Autonomous Vehicle, achieved zero collision out of 1 million episode of lane changes
• June 2017 - Aug 2017: Computational Science Summer Student Internship (NSF Mathematics Science Graduate Internship), Lawrence Berkeley National Laboratory
  • Developed code based on AMReX, which is a massively parallel adaptive mesh refinement framework, to solve wave equations in C++ and Fortran
  • Added node-level parallelism with a MPI plus Open MP approach. Improved thread scalability by experimenting the combination of MPI tasks and threads, achieved 1.5x speed up compared with pure MPI approach

Research experience

• Sep 2017 - Present: Research Assistant, Florida State University
  • Extracted and labeled 120000 vehicle trajectories with a total driven distance of 45000km
  • Trained Logistic Regression/Support Vector Machine/LSTM models to predict human intentions by learning from human driving dataset. Achieved 87.8% accuracy in predicting cooperative/adversarial behavior in lane change
  • Trained a Compact Support Neural Network (CSNN) to quantify uncertainty and detect long-tail rare events. CSNN obtained similar accuracy in classifying in-distribution samples compared with normal neuron based network, and achieved AUC of 0.99 in detecting out-of-distribution samples

Skills

• Machine Learnin (Pytorch, Tensorflow)
  • Deep Learning (CNN, RNN)
  • Reinforcement Learning
• CFA
• Programming (C++, Python, SQL)

Publications

Slat noise suppression based on upstream parallel mass injection

Slat noise suppression with mass injection

Playing Atari Ball Games with Hierarchical Reinforcement Learning

Talks

Indirect noise generation

March 18, 2017

Conference proceedings talk at SIAM Southeastern Atlantic Section Conference (SIAM-SEAS), Tallahassee, FL

Teaching

Calculus

Pre-Calculus

Service and leadership

• Fellows society at Florida State University