Mode-Constrained Exploration for Model-Based Reinforcement LearningAidan Scannell, Carl Henrik Ek, Arthur RichardsSep 24, 2022Code Source Document Followreinforcement-learning machine-learning gaussian-processes optimal-control robotics python TensorFlow GPflow researchPublicationsMode-constrained Model-based Reinforcement Learning via Gaussian ProcessesWe present a model-based RL algorithm that constrains training to a single dynamic mode with high probability. This is a difficult problem because the mode constraint is a hidden variable associated with the environment’s dynamics. As such, it is 1) unknown a priori and 2) we do not observe its output from the environment, so cannot learn it with supervised learning.Aidan Scannell, Carl Henrik Ek, Arthur RichardsPDF Cite Code Project Poster Source Document FollowEventsModel-based reinforcement learning under uncertaintyIn this talk I’ll present our recent paper that has just been accepted into AISTATS 2023, titled Mode-Constrained Model-Based …Feb 13, 2023 11:30 AM — 12:30 PM University of CambridgeAidan ScannellProject Slides
