ARLBench: Flexible and Efficient Benchmarking for Hyperparameter Optimization in Reinforcement Learning
Recent Posts
When Are RL Hyperparameters Benign? Disentangling Objective and Data Quality Permalink
TL;DR: AUsing offline goal-conditioned RL to isolate the effects of exploration and objective design, this study shows that hyperparameter landscapes can be ...
2024 in AutoRL
TL;DR: From integrating RL with VLMs and LLMs to hyperparameter tuning, environment design, and generalization, 2024 was packed with innovation. We’ve highl...
Dreaming of Many Worlds: Learning Contextual World Models Aids Zero-Shot Generalization Permalink
TL;DR: A new contextual world-model that helps to generalize better to new scenarios by understanding contextual factors like robot mass or strength. This co...