Viktor Losing

Papers

Total Citations

H-Index

About

Viktor Losing specializes in interactive machine learning, robotics, and human movement understanding, with a focus on developing adaptive systems that learn continuously in real-world environments. His most cited work, "Interactive online learning for obstacle classification on a mobile robot" (2015, 43 citations), introduces a novel architecture for incremental online learning in high-dimensional feature spaces. By leveraging learning vector quantization, Losing addresses the stability-plasticity dilemma—a fundamental challenge in continual learning—through adaptive insertion of representative vectors, enabling mobile robots to classify obstacles interactively without forgetting previously learned patterns. This contribution is pivotal for autonomous systems operating in dynamic settings. In his more recent project, "Machine learning for human movement understanding" (2020, 7 citations), Losing shifts focus to assistive robotics, aiming to develop technology that recovers and maintains human motor skills while extending the scope of human activity. His goal is to create systems that adapt to users' personal behavior patterns in real-time, fostering continuous collaboration between humans and robots. Losing's work bridges theoretical advances in online learning with practical applications in robotics and human-robot interaction, making him a notable figure in adaptive machine learning and assistive technology.