欢迎光临散文网会员登陆 & 注册

Reinforcement Learning_Code_Blackjack_Monte Carlo Learning

2023-03-25 18:13 作者:别叫我小红 0人读过 | 我要投稿

Blackjack.py

Visualization of reward and policy are are respectively shown below.

Fig. 1. Reward visualization.

Fig. 2. Policy Visualization with usable ace.

Fig. 3. Policy Visualization without usable ace.

The above codes are based on Gymnasium Documentation's tutorial "Solving Blackjack with Q-Learning", but solving Backjack with Monte Carlo learning.

[1] https://gymnasium.farama.org/tutorials/training_agents/blackjack_tutorial/

标签：

Reinforcement Learning_Code_Blackjack_Monte Carlo Learning的评论 (共条)