About me

I am currently a Research Associate under the AI Singapore Programme at the CPS Research Group, Nanyang Technological University (NTU). I possess in-depth expertise in machine learning research and algorithm development. My research interest includes:

• Reinforcement Learning
• Learning to Optimization
• Safety mechanisms of Machine Learning
• Robotics System

In addition to these core areas, I am eager to expand my research into emerging fields that align with my current expertise, such as generative AI, recommender system, data mining, federated learning, and related areas.

💻 Professional Experience

• Research Associate
  October 2022 - Present
  Arvind Easwaran, CPS Research Group, NTU, Singapore
  • ML safety navigation algorithm and robot control architecture for CPS system.
  • Automated assessment of trustworthiness for AI Training Programs (ATP).
  • Equipment setup, inventory organization, and facility maintenance.
    
    
• Student Researcher
  April 2022 - May 2022
  Guillaume SARTORETTI, MARMOT Lab, NUS, Singapore
  • Multi-agent Pathfinding research for warehouse systems.
  • Reinforcement learning-based path planner design.
  • Algorithm performance testing and data collection.
  • Comparison with baseline planners and results summarization.

🎓 Education

• Master of Science in Mechanical Engineering
  National University of Singapore, Singapore
  Completed: June 2022

• Bachelor of Science in Mechanical Engineering
  Nanjing University of Science And Technology, China
  Completed: July 2021

🔬 Research Experience

• AI Singapore: Automated assessment of trustworthiness for AI Training Programs
  Dec 2023 - Present
  Introduces a theoretical approach that leverages regret minimization within a teacher-student framework to provide immediate training feedback and fairness comparison. Allowing sample efficiency while preserving optimal solution.

• AI Singapore: ML safety navigation algorithm for CPS system
  Status: Under Review, Oct 2022 - Present
  Formulate a autonomous driving problem under the setting that is sensitive to the for safety configuration. This approach outperforms various baselines in terms of performance.

• Mechanical Engineering Project: Individual Voting for RL and Search-based Algorithm Combination in Multi-agent Pathfinding
  Aug 2021~Oct 2022
  Proposed an Individual Voting mechanism to optimize the pathfinding algorithm’s performance in dead/livelock situations. A decentralized RL algorithm and a centralized search-based algorithm are learned to combine.

• Final Year Project: Reinforcement Learning in Continuous Control Problem
  Sept 2020 - May 2021
  Implemented a RL-based algorithm for robust manipulator operation using RGB image inputs and Cartesian coordinate outputs. Achieved autonomous grasping for diverse workpiece structures.

🏆 Awards & Honors

• Outstanding Graduate Award in Mechanical Engineering, NJUST, China, 2021
• School of Mechanical Engineering School-level Scholarship, NJUST, China, Spring 2019-2020
• Second Prize in Jiangsu Province Mechanics Competition, China, 2019
• Outstanding Organizer of Literary Activities in Student Union, NJUST, China, Spring 2019
• School of Mechanical Engineering School-level Scholarship, NJUST, China, Spring 2018-2019
• School of Mechanical Engineering School-level Scholarship, NJUST, China, Fall 2018-2019
• School of Mechanical Engineering School-level Scholarship, NJUST, China, Spring 2017-2018

🛠️ Technical Skills

I am good at the skills below:

• Programming Languages: Python, C++, MATLAB, Origin
• Operation Platforms: Linux, ROS, Docker, Jetson Nano, Arduino, Raspberry Pi
• CAD Software: AutoCAD, SolidWorks
• Game Development: Unity, OpenAI Gym