Georg Krempl

Georg Krempl is Lecturer at Utrecht University's Department of Information and Computing Sciences. His research develops adaptive learning algorithms for data streams with concept drift and verification latency. He specializes in probabilistic active learning methods and statistical approaches to data stream mining.

Research Focus: Krempl's work addresses challenges in evolving data environments, particularly developing robust classification under delayed label scenarios and concept drift. Recent projects include generative adversarial networks for domain adaptation and Bayesian methods for anticipatory classification.

Professional Activities: Organizer of the Interactive Adaptive Learning workshop series (ECML PKDD) and editorial board member for multiple machine learning journals. His group focuses on both theoretical foundations and practical applications in data stream mining.