Minjong Yoo
Papers
2
Total Citations
5
H-Index
2
About
Minjong Yoo is a rising researcher in artificial intelligence, specializing in reinforcement learning (RL) and multi-task decision-making. His work focuses on enabling AI agents to learn efficiently from offline datasets and adapt to new tasks without additional training. In his highly cited paper "Skills Regularized Task Decomposition for Multi-task Offline Reinforcement Learning" (2024, 3 citations), Yoo introduced a novel framework that leverages common skills across diverse tasks to improve data efficiency and generalization in offline RL—a critical advancement for real-world applications where online interaction is costly or unsafe. His second major contribution, "SemTra: A Semantic Skill Translator for Cross-Domain Zero-Shot Policy Adaptation" (2024, 2 citations), pioneers the use of semantically interpretable skill representations to enable zero-shot policy transfer across different domains. This work allows AI systems to interpret user commands in multi-modal snippets and execute long-horizon tasks in entirely new environments without prior exposure. Yoo’s research bridges the gap between skill-based learning and cross-domain adaptation, offering practical solutions for robotics and autonomous systems. His achievements demonstrate a clear trajectory toward building more flexible, generalizable, and human-interpretable AI agents.
Research Focus
Key Achievements
Top Papers
- 1
- 2