Cliff World

Concept

A grid world environment with a dangerous cliff region, used to compare the policies learned by SARSA (on-policy) and Q-learning (off-policy).

Mentioned in 1 video

Videos Mentioning Cliff World

Stanford AA203 Optimal and Learning-Based Control | Spring 2026 | Lecture 17: RL Value-Based Methods

Stanford AA203 Optimal and Learning-Based Control | Spring 2026 | Lecture 17: RL Value-Based Methods

Stanford Online

A grid world environment with a dangerous cliff region, used to compare the policies learned by SARSA (on-policy) and Q-learning (off-policy).