Linji Wang — AI robotics engineer specializing in curriculum learning and embodied reinforcement learning
open classic →
Computer Science Ph.D. researcher at George Mason's RobotiXX Lab, focused on automatic curricula, adaptive rewards, and reliable policy inference.
A teacher generates training tasks from online learner performance while grounding the curriculum in limited reference samples from the target task distribution.
- +6.8% success — wheeled navigation in constrained environments
- +6.1% success — quadruped locomotion in confined 3D spaces
A teacher re-weights auxiliary rewards as the robot's proficiency grows — training wheels that fade away on their own.
- Simulation: +122.62% off-road mobility; same threshold 3× faster
- Physical: 5/5 trials vs 2/5 for expert-designed rewards
Plan with full robot dynamics near the robot, and progressively simpler dynamics along the horizon — full fidelity exactly where it matters.
- Improved performance across three augmented planners
- RobotiXX placed 2nd in both simulation and physical phases
A VR data-collection and evaluation framework that captures whole-body motion in procedurally generated clutter and automatically retargets it to a humanoid model.
- 348 trajectories across 145 diverse 3D scenes
- Benchmarks stability, collision safety, and scene clutter
Who am I
I am an AI robotics and systems engineer and Computer Science Ph.D. researcher at George Mason University's RobotiXX Lab, advised by Dr. Xuesu Xiao. I build automatic curriculum-learning and reinforcement-learning systems for embodied robots — from GPU-parallel simulation and physical off-road validation to humanoid locomotion benchmarking and policy inference. My systems work also includes Aurora PostgreSQL query-performance and compatibility engineering at AWS.