Skip to main content

A reinforcement learning environment based off the board game Risk

Project description

A reinforcement learning environment based off the board game Risk. Designed with reinforcement learning in mind, this project aims to streamline research for economy-based games.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Dependencies

  • numpy <= 1.14
  • pygame <= 1.9

User installation

The easiest way to install this package is through PyPi

pip install rlrisk

Otherwise you can manually install this package by cloning the repository to your local computer

git clone https://github.com/civrev/RLRisk

Then navigate to that directory and run setup.py

python setup.py install

Working with RLRisk

RLRisk was a project concieved to streamline research by coding an environment already friendly to reinforcement learning techniques. As such working with RLRisk is extremely easy.

The Environment

The environment is the standard Wolrd Domination game rules for Risk by Hasbro. You can find these rules here: https://www.hasbro.com/common/instruct/risk.pdf

The environment is located in the file risk.py and is implemented using the class Risk(). The class is flexible enough to be initiallized with extremely custom rules, but for the most common game of Risk you can create a Risk object using only a list of players

from rlrisk.agents import *
from rlrisk.environment import *

players = [BaseAgent(), AggressiveAgent()]
env = Risk(players)
results = env.play()

Agents

RLRisk comes with 3 agents, the BaseAgent, Human, and AggressiveAgent classes. All new agents must be subclasses of the BaseAgent class, but looking at the BaseAgent will show you that the framework of agents for RLRisk is very straight forward. It takes in the information it need to make a decision, and then it outputs a decision.

Final Remarks

This is my Senoir Project for my B.S. in Computer Science at the University of North Georgia (Graduating May, 2018)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
rlrisk-1.0.2.3-py3-none-any.whl (984.3 kB) Copy SHA256 hash SHA256 Wheel py3 Apr 25, 2018

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page