Tabular Q-learning on OpenAI Gym's Frozen Lake. Samples from the observation space, updating the Q-value of each state/action pair. Starts by exploring the observation space through taking random actions, then over time exploits the known Q-values by taking the argmax at the current state.
-
Notifications
You must be signed in to change notification settings - Fork 1
yimothysu/frozen-lake
About
Tabular Q-learning on OpenAI Gym's Frozen Lake.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published