Pascal Klink (TU Darmstadt): Self-paced Reinforcement Learning

Loading Events
  • This event has passed.

Pascal Klink (TU Darmstadt): Self-paced Reinforcement Learning

Thursday Morning Lectures
Abstract:
Generalization and adaptation of learned skills to novel situations is a core requirement for intelligent autonomous robots. Although contextual reinforcement learning provides a principled framework for learning and generalization of behaviors across related tasks, it generally relies on uninformed sampling of environments from an unknown, uncontrolled context distribution, thus missing the benefits of structured, sequential learning. We introduce a novel relative entropy reinforcement learning algorithm that gives the agent the freedom to control the intermediate task distribution, allowing for its gradual progression towards the target context distribution. Empirical evaluation shows that the proposed curriculum learning scheme drastically improves sample efficiency and enables learning in scenarios with both broad and sharp target context distributions in which classical approaches perform sub-optimally.
 
Bio:
Pascal is a Ph.D. student at the Intelligent Autonomous Systems (IAS) Group at TU Darmstadt. At IAS, he works for the ROBOLEAP project, where he develops methods for reinforcement learning in unstructured, partially observable real world environments. Before starting his PhD, Pascal completed his Bachelor’s degree in Computer Science and Master’s degree in Autonomous Systems at the TU Darmstadt. Within his Master’s thesis he worked on “Generalization and Transferability in Reinforcement Learning” and was supervised by Hany Abdulsamad, Boris Belousov and Jan Peters

Event Details

Date: February 20, 2020 @ 10:00 am - 11:00 am CET
Time: 10:00 am - 11:00 am
Venue: MAR23 4.064
Address: Marchstraße 23