Haohong Lin

Papers

Total Citations

H-Index

About

Haohong Lin is a rising researcher at the intersection of robotics, reinforcement learning, and safe AI, whose work tackles two of the field’s most pressing challenges: enabling robots to perform complex, long-horizon manipulation tasks and ensuring that learning algorithms operate safely. Lin’s most notable contribution is the **Tactile Ensemble Skill Transfer (TEST)** framework, a pioneering offline reinforcement learning approach that leverages tactile feedback to generalize robotic skills for furniture assembly—a notoriously difficult problem due to its non-repetitive, multi-step nature. This work, published in 2024, has already garnered 6 citations, signaling its immediate impact on the robotics community. Complementing this, Lin co-developed a comprehensive **benchmarking suite for offline safe reinforcement learning** (2023, 5 citations), providing standardized datasets and evaluation protocols that are essential for advancing safety-critical learning algorithms. This dual focus—on both capability and safety—positions Lin as a key contributor to the next generation of autonomous systems. With a clear trajectory toward bridging tactile sensing and reinforcement learning, Haohong Lin’s work is shaping how robots learn to interact with the physical world reliably and safely.