Marek Herde

Papers

Total Citations

H-Index

About

Marek Herde is a researcher at the forefront of machine learning and robotics, specializing in active learning, human-in-the-loop systems, and probabilistic classification. His work addresses a critical industrial challenge: enabling robots to autonomously adapt to new tasks—such as object sorting—without costly manual reprogramming. Herde’s major contributions include developing probabilistic active learning frameworks that allow robots to intelligently query the most informative training examples, dramatically reducing the data needed for effective learning. In his most-cited work, “Active Sorting – An Efficient Training of a Sorting Robot with Active Learning Techniques” (2018, 6 citations), he demonstrated how active learning can streamline robot training for industrial automation. Herde also advanced label-quality improvement by introducing methods to combine self-reported confidences from multiple uncertain annotators (2019, 4 citations), a novel approach that leverages annotator uncertainty to enhance data reliability. His research on automated active learning with robots (2018, 2 citations) further explores intuitive human-robot interaction, making learning processes transparent and accessible. With a focus on bridging theoretical active learning with practical robotic applications, Herde’s work is paving the way for more flexible, efficient, and autonomous industrial systems.