Teng‐Yok Lee

Mitsubishi Electric (Japan), Boston University

Papers

2

Total Citations

25

H-Index

2

About

Teng-Yok Lee is a researcher at the forefront of visual analytics and robotic manipulation, bridging the gap between complex AI systems and practical, real-world control. His work primarily focuses on two challenging domains: making deep reinforcement learning (RL) interpretable for physical tasks, and enabling robots to perform semantically-aware object manipulation. In his highly cited 2020 work, "DynamicsExplorer," Lee tackled the "black box" problem of deep RL, creating a visual analytics system that allows engineers to understand and debug the control policies of robots trained for dynamic tasks. This contribution is critical for deploying RL in safety-critical environments, earning 22 citations for its novel approach to human-in-the-loop analysis. Lee also advanced the field of robotic placement with his 2019 study on "Pose-Aware Placement," which integrated semantic labels—like brand names—to guide a dual-arm robot in placing objects with proper orientation and context. This work moves beyond simple pick-and-place toward more intelligent, context-driven manipulation. Through these efforts, Lee is helping to build the foundational tools and interfaces needed for the next generation of autonomous, explainable robotic systems.

Research Focus

Key Achievements

2
H-Index
2
Papers
25
Total Citations
13
Avg Citations/Paper
🏆 Most Cited Paper
DynamicsExplorer: Visual Analytics for Robot Control Tasks involving Dynamics and LSTM-based Control Policies
22 citations · 2020
📈 Most Prolific Year: 2020 (1 Papers)
🤝 Key Collaborators: 13
🏛 Institutions: Mitsubishi Electric (Japan), Boston University

Top Papers

  1. 1
  2. 2

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago