Skip to content

hsezhiyan/Easy21_Reinforcement-Learning

Repository files navigation

Easy 21 Game Implementation using Reinforcement Learning

This assignment was completed for David Silver's Reinforcement Learning course.

Game Definition

prog1.py defines the rules of the game, which is a slight variant of blackjack.

Monte Carlo Method

prog2.py implements the Monte Carlo methods for calculating the value of the state. The state is defined by two parameters, the sum of the cards the dealer holds and the sum of the cards the player holds. The value of a state is a float contained in (-1.0, 1.0).

To run this file: python prog2.py


Usage Data

SARSA Method

prog3.py implements the SARSA method, also for calculating the value of the state.

To run this file: python prog3.py


Usage Data

About

Easy 21 Reinforcement learning assignment, David Silver's course.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages