Max Robotics — Global Service Robot Directory + US Certification

About

Keting Lu is a researcher at the forefront of bridging reinforcement learning (RL) with knowledge representation and reasoning (KRR) for autonomous robotics. Her work addresses a fundamental challenge: how to equip RL agents—which excel at learning from interaction but lack declarative reasoning—with the structured knowledge needed for complex tasks. In her highly cited 2018 paper, Lu pioneered methods for robots to represent and reason with knowledge extracted from RL, effectively merging data-driven learning with symbolic AI. Her subsequent research on Automated Experience Grafting (AutoEG, 2020) tackles the sample inefficiency of deep RL by enabling agents to learn from low-quality early trials, significantly accelerating policy learning. Lu has also developed integrated frameworks for robot dialog and navigation, combining probabilistic reasoning with RL to enhance task completion in real-world environments. With over 25 citations across her core publications, her contributions are shaping the next generation of intelligent systems that can both learn from experience and reason with knowledge—a critical step toward truly autonomous robots.