Vos, Thomas (2022) Reinforcement Learning and Evolutionary algorithms in the Stochastic Environment of Blackjack. Bachelor's Thesis, Artificial Intelligence.
|
Text
BachelorThesiss3162443ThomasVos.pdf Download (810kB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (120kB) |
Abstract
The casino game blackjack requires player’s actions to play. These actions have an optimal strategy which we learn with two different reinforcement learning (RL) algorithms (Q-learning and QV-learning) and 2 different evolutionary algorithms (genetic algorithm (GA) and particle swarm optimization (PSO)). The maximum win rate for blackjack is around 42,5%, and our best performing algorithm (QV-learning) achieved a 42,47% win rate with the best exploration policy. Q-learning and GA both achieved a 42,30% win rate with their best strategies. Whereas PSO performed badly only achieving 21.2% which indicates it is not suited for the problem.
| Item Type: | Thesis (Bachelor's Thesis) |
|---|---|
| Supervisor name: | Sabatelli, M. |
| Degree programme: | Artificial Intelligence |
| Thesis type: | Bachelor's Thesis |
| Language: | English |
| Date Deposited: | 01 Jul 2022 12:20 |
| Last Modified: | 01 Jul 2022 12:20 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/27515 |
Actions (login required)
 |
View Item |