Barebones blackjack playing agent with reinforcement learning. Created for COSC 6342: Machine Learning. Code adapted from one of Jeremy Zhang's RL implementations and will be posted post-submission.
- Exploration rate vs. result count
- Learning rate vs. result count
- Play vs. dealer results
Rounds | Wins | Losses | Draws |
---|---|---|---|
1k | 444 | 164 | 392 |
10k | 4062 | 1604 | 4334 |
100k | 41181 | 15975 | 42844 |