Kenji Tei
Papers
7
Total Citations
56
H-Index
5
About
Kenji Tei is a leading researcher in self-adaptive systems, focusing on how software can intelligently adjust its behavior in dynamic, unpredictable environments. His work centers on integrating reinforcement learning (RL) and formal methods to create systems that learn, adapt, and guarantee their own performance. Tei’s major contributions include pioneering a runtime monitoring framework that enforces safety invariants on RL agents exploring complex environments (16 citations), and developing a meta-RL approach that enables self-adaptive systems to learn high-performance adaptation policies without prior environmental knowledge (10 citations). He has also advanced the practical application of these ideas, demonstrating a real-world self-adaptive robot using discrete controller synthesis (5 citations). His research on preference adaptation (11 citations) and iterative requirement relaxation (9 citations) addresses the critical challenge of balancing multiple, often conflicting, user needs at runtime. Tei’s work is notable for bridging the gap between theoretical adaptation models and real-world deployment, making self-adaptive systems more robust, user-aware, and scalable.
Research Focus
Key Achievements
Top Papers
- 1
- 2Preference Adaptation: user satisfaction is all you need!11 citations · 2023
- 3A Meta Reinforcement Learning-based Approach for Self-Adaptive System10 citations · 2021
- 4
- 5
- 6
- 7