Sotirios Chatzis
Papers
3
Total Citations
112
H-Index
3
About
Dr. Sotirios Chatzis is a leading researcher in statistical machine learning, with a particular focus on Bayesian nonparametrics, variational inference, and their applications to robotics and artificial intelligence. His major contributions include pioneering variational Bayesian methodologies for hidden Markov models, where he introduced Student's-t mixtures to enhance robustness against outliers—a foundational work that has garnered 62 citations. In the realm of robotics, Dr. Chatzis has advanced robot learning by demonstration through innovative nonparametric Bayesian approaches, enabling more flexible and data-efficient imitation learning. His quantum-statistical framework for robot learning, which reimagines Gaussian mixture regression through the lens of quantum mechanics, represents a novel interdisciplinary contribution with 22 citations. Collectively, his work bridges theoretical rigor with practical deployment, offering scalable solutions for autonomous systems. With over 100 citations across his most influential papers, Dr. Chatzis continues to shape how machines learn from human demonstration, pushing the boundaries of Bayesian statistics and its real-world impact.
Research Focus
Key Achievements
Top Papers
- 1
- 2A nonparametric Bayesian approach toward robot learning by demonstration28 citations · 2012
- 3A Quantum-Statistical Approach Toward Robot Learning by Demonstration22 citations · 2012