Papers
5
Total Citations
59
H-Index
4
About
Jieke Shi is a researcher at the forefront of multi-agent systems and trustworthy reinforcement learning. His work bridges the gap between theoretical algorithms and real-world deployment, with a particular focus on making autonomous systems both efficient and secure. Shi’s foundational contributions include an auction-based rescue task allocation approach for heterogeneous multi-robot systems, which has garnered 38 citations and remains a key reference in the field. He has also advanced real-time, decentralized task scheduling for multi-agent systems under continuous damage, addressing critical robustness challenges. More recently, Shi has turned his attention to the safety and reliability of sequential decision-making processes (SDPs), which underpin technologies like autonomous driving and robotic control. His work on curiosity-driven testing introduces novel methods for uncovering vulnerabilities in deep learning-based SDPs. Additionally, his research on backdoor attacks in offline reinforcement learning—detailed in the BAFFLE framework—has exposed significant security risks in data-sharing paradigms, earning 8 combined citations. By tackling both performance and trustworthiness, Jieke Shi is shaping the next generation of resilient, intelligent autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Curiosity-Driven Testing for Sequential Decision-Making Process6 citations · 2024
- 4Baffle: Hiding Backdoors in Offline Reinforcement Learning Datasets5 citations · 2024
- 5BAFFLE: Hiding Backdoors in Offline Reinforcement Learning Datasets3 citations · 2022