Skip to content
This repository has been archived by the owner on Mar 3, 2018. It is now read-only.

victorgomes/RL-Framework

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

This framework was created in order to compare different 
Reinforcement Learning algorithms. 

The key principle here is to encapsulate what really changes 
(the agent and the environment), so later one can alter it 
without affect what it doesn’t change (the interaction between 
the agent and the environment.

The framework is quite simple, there are just two packages. 
The first one, tools, contains all functions that one may need 
to develop a RL algorithm, like the value functions, 
V (s) and Q(s, a); the transition function, T (s, a, s); 
and the reward function, R(s, a). 
Those classes model a fuction that use the State and Action class, 
which contains an integer identifier, as parameters and return a double.

The second package, main, contains the keys of this framework: 
two interfaces, IAgent and IEnvironment.

To run the project:
javac Main.java
java Main

About

Reinforcement Learning Framework

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages