Reinforcement Learning

Implementation of a simple Q-Learning algorithm that learns from interacting with an environment.

How to use

Best use with Clion

• run WorldDomination for game
  • you can change the amount of games the agent will play by changing the parameter number of fit method in main
  • you can change the strategy in main when creating the agent by either passing new SoftMaxPolicy() or new ThresholdPolicy(double explorationrate). Differences were explained in the presentation (see PowerPoint file in project folder)
• run WD_Test for tests

Documentation

https://ob-algdatii-ss18.github.io/leistungsnachweis-firststeptoworlddomination/