Wonjoon Goo
Papers
3
Total Citations
37
H-Index
2
About
Wonjoon Goo is a researcher advancing the frontiers of robot learning, with a focus on imitation learning, task generalization, and efficient human-to-robot skill transfer. His most influential work, "Better-than-Demonstrator Imitation Learning via Automatically-Ranked Demonstrations" (2019, 32 citations), tackles a fundamental limitation of traditional imitation learning: the learner’s performance is typically capped by the demonstrator’s skill. Goo introduces a method that automatically ranks demonstrations, enabling robots to surpass their teacher’s ability without requiring explicit human preference labels—a breakthrough for scalable, real-world deployment. In related work, Goo explores learning multi-step robotic tasks from observation (2018) and one-shot generalization using auxiliary video for activity localization (2019), addressing the data inefficiency and ambiguity that plague demonstration-based programming. His contributions are especially valuable for students and researchers seeking to build robots that learn quickly, generalize robustly, and improve beyond their training data. Goo’s research sits at the intersection of machine learning, robotics, and human-robot interaction, offering practical pathways toward more autonomous and capable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2Learning Multi-Step Robotic Tasks from Observation.3 citations · 2018
- 3