Skip to content

Four multiagent learning algorithms for the Markov game "Soccer"

Notifications You must be signed in to change notification settings

xh2310/MultiAgentQLearning

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

MultiAgentQLearning

This solution implements four multiagent learning algorithms to be used with the "Soccer" zero-sum, two-player, non-deterministic Markov game.

  1. Plain Q-Learning
  2. Friend-Q
  3. Foe-Q
  4. Correlated-Q

Prerequisites:

  1. Windows 7 or later and .NET 4.5.2
  2. Microsoft Solver Foundation 3.0 Express or higher (https://msdn.microsoft.com/en-us/library/ff524499(v=vs.93).aspx)

To run: Compile the provided solution using MSBuild or Visual Studio and run the resulting executable. Follow the instructions to run one of the algorithms.

About

Four multiagent learning algorithms for the Markov game "Soccer"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages