Evangelos Kokkevis
Papers
2
Total Citations
78
H-Index
2
About
Evangelos Kokkevis is a computer scientist whose pioneering work in machine learning and robotics has left a lasting mark on how autonomous systems perceive and navigate their environments. His research centers on computational learning theory, particularly the inference of finite automata with stochastic output functions—a method that enables machines to build probabilistic models of complex, uncertain worlds from limited data. Kokkevis’s most cited paper, “Inferring Finite Automata with Stochastic Output Functions and an Application to Map Learning” (1995), has garnered over 50 citations, demonstrating its foundational influence on fields such as robot mapping, spatial reasoning, and adaptive control. By showing how automata can be learned from noisy sensor streams, he provided a rigorous framework for robots to construct internal maps without explicit programming—a key step toward autonomous exploration. His work bridges theoretical computer science and practical robotics, inspiring subsequent research in probabilistic modeling and reinforcement learning. Kokkevis’s contributions remain a touchstone for students and engineers seeking to understand how machines can learn structure from stochastic environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2