BeNeRL Reinforcement Learning Seminar: Daniel Palenicek from TU Darmstadt

Join us for the next BeNeRL Reinforcement Learning Seminar with speaker Daniel Palenicek, a PhD student from TU Darmstadt. Daniel will be presenting on ‘Sample Efficiency in Deep RL: Quo Vadis?’. The seminar will take place on September 12, 16.00-17.00 (CET) via Zoom.

Daniel’s talk will present two studies that challenge conventional wisdom in deep RL, offering fresh perspectives on accelerating RL algorithms and highlighting some fundamental limitations. The first study explores the limits of value expansion methods in model-based RL, and the second study introduces CrossQ, a novel approach that dramatically improves sample efficiency in off-policy RL.

Daniel is a PhD student at the Intelligent Autonomous System Group, TU Darmstadt, where he is advised by Prof. Jan Peters. His research lies at the intersection of reinforcement learning and robotics, focusing on increasing sample efficiency and scaling model-free and model-based reinforcement learning algorithms.

Join the Zoom meeting here.

Learn more about the talk and speaker on the BeNeRL website.