Erik Englesson

Erik Englesson is a Researcher at the Division of Robotics, Perception and Learning at KTH Royal Institute of Technology. His work is supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP-AI/MLX). He holds a PhD from KTH Royal Institute of Technology, focusing on label noise in image classification from an aleatoric uncertainty perspective.

His research interests center on robustness and uncertainty quantification in deep learning, particularly addressing label noise and combining aleatoric and epistemic uncertainties. He teaches Probabilistic Graphical Models (DD2420) at KTH.

Recent publications explore topics like gradient-based explanation methods, autoencoder parameterization, and noise-robust classification strategies. His work frequently intersects with computer vision and theoretical deep learning principles.

Englesson collaborates with prominent researchers like Hossein Azizpour and receives funding from strategic initiatives in AI and autonomous systems. His research bridges foundational theory with practical applications in healthcare imaging and robust model design.