Georgios Leontidis

Papers

Total Citations

H-Index

About

Georgios Leontidis is a leading researcher in artificial intelligence and machine learning, with a primary focus on Bayesian deep learning, self-supervised learning, and their applications in medical imaging and computer vision. His most-cited work, "Deep Bayesian Self-Training" (2020, 36 citations), introduces a novel framework that combines Bayesian uncertainty estimation with self-training, enabling more robust and reliable model predictions in data-scarce scenarios—a critical advancement for domains like healthcare where labeled data is limited. Leontidis has made significant contributions to developing uncertainty-aware AI systems, addressing key challenges in model calibration and out-of-distribution detection. His research has been widely recognized for bridging theoretical machine learning with practical, high-stakes applications, particularly in medical image analysis where his methods improve diagnostic accuracy and trustworthiness. Beyond his technical contributions, Leontidis is known for fostering collaborative research, as evidenced by his acknowledgments of colleagues who manually annotated datasets for his studies. His work continues to influence the next generation of AI researchers, emphasizing the importance of robust, interpretable, and uncertainty-quantified models in real-world deployment.