ConceptAI & ML
Stability in reinforcement learning
Stability refers to whether an algorithm converges reliably without requiring excessively delicate hyperparameter tuning or suffering large performance degradations.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Hyperparameters1Reinforcement learning algorithms1
Lecturers
1Connections
2The lectures that define it or assume it, and the concepts next to it.