Jiayi Tan
Papers
2
Total Citations
4
H-Index
2
About
Jiayi Tan is a rising researcher in robotics and artificial intelligence, specializing in efficient policy learning for complex manipulation tasks. Her work addresses a critical bottleneck in deep reinforcement learning: how to train robots to perform sophisticated manipulations using only a limited number of human demonstrations. Tan’s major contribution lies in developing algorithms that maximize learning from sparse, imperfect data. Her 2025 paper, "E-GAIL: efficient GAIL through including negative corruption and long-term rewards for robotic manipulations," introduces a novel framework that enhances Generative Adversarial Imitation Learning by incorporating negative corruption and long-term reward signals, achieving high efficiency with minimal demonstrations. This builds on her earlier 2023 work, "Learning Complicated Manipulation Skills Via Deterministic Policy with Limited Demonstrations," which tackled the challenge of mismatched human-robot demonstration quality. Though early in her career, Tan’s research has already garnered citations, signaling growing impact in the robotics community. Her focus on practical, data-efficient solutions positions her work as foundational for advancing autonomous robotic systems in real-world settings.
Research Focus
Key Achievements
Top Papers
- 1
- 2