Andre He
Papers
2
Total Citations
16
H-Index
2
About
Andre He is a leading researcher in scalable robot learning, with a focus on bridging the gap between data-driven methods and real-world robotic manipulation. His major contributions center on creating large-scale, diverse datasets and developing semi-supervised techniques to enable robots to follow natural language instructions. He is the lead author of "BridgeData V2" (2023, 12 citations), a landmark dataset containing over 60,000 robotic manipulation trajectories across 24 environments, designed to democratize robot learning by using low-cost hardware. This work has become a foundational resource for the field, accelerating research on generalization and transfer in robotics. He also pioneered semi-supervised language interfaces in "Goal Representations for Instruction Following" (2023, 4 citations), addressing the critical challenge of learning from limited labeled data—a key bottleneck in instruction-following robots. His work is notable for its practical impact, enabling robots to interpret commands like "put the towel next to the microwave" without requiring massive, costly annotations. Andre He’s research is shaping the future of accessible, scalable robot learning, making him a rising star in the robotics community.
Research Focus
Key Achievements
Top Papers
- 1BridgeData V2: A Dataset for Robot Learning at Scale12 citations · 2023
- 2