Challenges in Hyperparameter Optimization for Reinforcement Learning - Understanding AutoRL Through AutoML
Explore a 32-minute AutoML Seminar presentation where Theresa Eimer delves into the complexities of hyperparameter optimization in Reinforcement Learning (RL) and its relationship with Automated Machine Learning (AutoML). Learn how AutoML tools can be leveraged to better understand RL algorithm behaviors, particularly focusing on hyperparameter influence on training outcomes. Discover insights from ICML 2023-accepted research that investigates RL hyperparameters and establishes groundwork for improved AutoRL methods. Gain valuable understanding of the current challenges in Automated Reinforcement Learning, its development compared to broader AutoML applications, and potential future directions for integrating AutoML concepts into AutoRL frameworks.