Curriculum Vitae
Feiyang Wu
ML PhD student at Georgia Tech
My research interests lie at the intersection of optimization, reinforcement learning, and robotics. I develop algorithms for data-efficient robot learning, ideally with theoretical guarantees, and work across both reinforcement learning theory and embodied robot learning systems.
This page is the text-rendered version of my CV. The original PDF is available through the button above.
Education
Georgia Institute of Technology
Doctor of Philosophy in Machine Learning Expected 2027
Master of Science in Computer Science 2023
Chinese University of Hong Kong, Shenzhen
Bachelor of Engineering in Computer Science and Engineering 2020
Publications
Reinforcement Learning
arXiv preprint, 2026
arXiv preprint, 2026
ROBOTIS Best Paper Award (1st place) — CompRobotis, IROS 2026 (workshop version)
arXiv preprint, 2026
ICML 2026 Oral (0.7% acceptance rate)
ICRA 2026
SEEC: Stable End-Effector Control with Model-Enhanced Residual Learning for Humanoid Loco-Manipulation
ICRA 2026
★ Learn to Teach: Sample-Efficient Privileged Learning for Humanoid Locomotion over Real-World Uneven Terrain
RA-L 2025, ICRA 2026 Oral
Mathematics of Operations Research 2025
Infer and Adapt: Bipedal Locomotion Reward Learning from Demonstrations via Inverse Reinforcement Learning
ICRA 2024
Computer Vision
IEEE Transactions on Visualization and Computer Graphics (TVCG) 2020
Research and Working Experience
Research Intern
Georgia Tech Research Institute
- Built a tele-operation data collection platform in Isaac Lab and Isaac Sim for UR5 manipulation across a range of crafted scenes.
- Collected data, augmented it with generative techniques, trained DP3 and diffusion policies, and connected Isaac Lab with ROS2 for hardware demonstration.
Linear Programming Solver Developer
Shenzhen Research Institute of Big Data, CUHK Shenzhen
- Led development for large-scale general-purpose linear programming solvers.
- Implemented primal-dual simplex variants, direct linear solves via Cholesky decomposition, and regularization and perturbation techniques for interior-point methods.
- Pushed computational performance toward the level of leading commercial solvers.
Research Intern
Didi Chuxing
- Worked on the on-device AR navigation system in the Visual Computing Group, improving pick-up route guidance through visual SLAM and cloud-assisted infrastructure.
- Developed and tested navigation features, analyzed user data, and built machine learning models including gradient boosting, LSTMs, and SVMs to improve system accuracy.
Skills
Programming Languages
C/C++, Python, Java, Objective-C, JavaScript, HTML/CSS, C#
Software and Tools
LaTeX, MATLAB, Maya, MeshLab, MS Office
Frameworks
PyTorch, TensorFlow, OpenCV, Scikit-learn, Armadillo