Implemented agent which chooses to play winning, losing or random moves by solving a simplified MDP model of the game using policy iteration.

Portions of MDP/solver code by Ciaran O'Reilly and Ravi Mohan used under MIT license.

test/model/mdp/ValueIterationSolverTest.java

@@ -0,0 +1,26 @@
+package model.mdp;
+
+import static org.junit.Assert.assertTrue;
+import model.mdp.MDP.MODE;
+
+import org.junit.Test;
+
+public class ValueIterationSolverTest {
+
+	@Test
+	public void testSolve() {
+		MDPSolver solver = new ValueIterationSolver();
+		
+		//solve for a score of 25 in at most 35 turns
+		int maxScore = 6;
+		int maxTurns = 10;
+		
+		MDP mdp = new MDP(maxScore,maxTurns,MODE.CEIL);
+		Policy policy = solver.solve(mdp);
+		
+		assertTrue(policy.size() >= maxScore);
+		assertTrue(policy.size() <= maxTurns);
+		
+		System.out.println("Policy: " + policy);
+	}
+}