Base Classes for the Java Reinforcement Learning Interface:
-----------------------------------------------------------
Action.java
ActionResult.java
Agent.java
Environment.java
Simulation.java
State.java

Below are two example uses of these base classes.  Many other implementations have been developed by Prof. Neller and the Gettysburg College students of CS 391 (Spring '03).  Contact Prof. Neller for more information/examples.

Additional Classes for the Heirarchical and Simple GridWorld Examples:
----------------------------------------------------------------------
GridActIterator.java
GridAction.java
GridEnvir.java
GridSim.java
GridState.java
GridStateIterator.java
SimpleGridAgentQL.java
SimpleGridAgentSARSA.java
TDATabQLGrid.java
TDATabSARSAGrid.java
TDAgent.java
TDAgentTab.java
TDAgentTabQL.java
TDAgentTabSARSA.java

Additional Classes for the Mountain Car Problem:
------------------------------------------------
DoubleArrayState.java
Featurizer.java
MtnCarAction.java
MtnCarAgentLGDSarsaTCM.java
MtnCarEnvironment.java
MtnCarSimulation.java
MtnCarState.java
TileCodingMemory.java
(Note: This is an adaptation of the C implementation by Rich Sutton and Sridhar Mahadevan, available at http://www-anw.cs.umass.edu/rlr/domains.html)