Niclas Pokel

Bio

Originally from Leipzig, I have pursued studies in Engineering, Mathematics, and Robotics across Munich, Stuttgart, Copenhagen, London and Zurich. My academic journey during my masters has covered a broad spectrum, ranging from probabilistic ensemble learning and learning theory to time series forecasting for electricity markets. Before joining the Grewe Lab, I gained industry experience in software development and data science, where I optimized LLMs for customer interactions, and in systematic trading within energy markets. At the Grewe Lab, my primary research focuses on continual learning, parameter efficiency, and bio-plausible learning, with a particular emphasis on compositionality. I originally joined the lab for my Master’s thesis on Automatic Speech Recognition (ASR) for impaired speech. I continue to actively drive this research area: beyond improving ASR models, we are now developing authoring tools specifically designed for speech therapy applications. Outside the lab, I am passionate about rock climbing, (technical) mountaineering and economic theory. If you are interested in my work, looking for an exchange, or exploring a potential collaboration, please don’t hesitate to reach out! Current opportunities for Master’s theses or student projects can be found in the Join the Team section of this website.