Changeset 11727 for branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Problems.GrammaticalOptimization.Test/TestBanditPolicies.cs

Timestamp:

12/29/14 11:02:36 (9 years ago)

Author:

gkronber

Message:

#2283: worked on grammatical optimization problem solvers (simple MCTS done)

File:

• branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Problems.GrammaticalOptimization.Test/TestBanditPolicies.cs (modified) (2 diffs)

branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Problems.GrammaticalOptimization.Test/TestBanditPolicies.cs

@@ -15 +15 @@
       var seedForPolicy = globalRand.Next();
       var nArms = 10;
+      //Console.WriteLine("Exp3 (gamma=0.01)");
+      //TestPolicyBernoulli(globalRand, nArms, new Exp3Policy(new Random(seedForPolicy), nArms, 1));
+      //Console.WriteLine("Exp3 (gamma=0.05)");
+      //estPolicyBernoulli(globalRand, nArms, new Exp3Policy(new Random(seedForPolicy), nArms, 1));
+      Console.WriteLine("Thompson (Bernoulli)");
+      TestPolicyBernoulli(globalRand, nArms, new BernoulliThompsonSamplingPolicy(new Random(seedForPolicy), nArms));
       Console.WriteLine("Random");
-      TestPolicyBernoulli(globalRand, nArms, new RandomPolicy(new Random(seedForPolicy), 10));
+      TestPolicyBernoulli(globalRand, nArms, new RandomPolicy(new Random(seedForPolicy), nArms));
       Console.WriteLine("UCB1");
-      TestPolicyBernoulli(globalRand, nArms, new UCB1Policy(10));
+      TestPolicyBernoulli(globalRand, nArms, new UCB1Policy(nArms));
       Console.WriteLine("UCB1Tuned");
-      TestPolicyBernoulli(globalRand, nArms, new UCB1TunedPolicy(10));
+      TestPolicyBernoulli(globalRand, nArms, new UCB1TunedPolicy(nArms));
       Console.WriteLine("UCB1Normal");
-      TestPolicyBernoulli(globalRand, nArms, new UCBNormalPolicy(10));
+      TestPolicyBernoulli(globalRand, nArms, new UCBNormalPolicy(nArms));
       Console.WriteLine("Eps(0.01)");
-      TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.01));
+      TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.01));
       Console.WriteLine("Eps(0.05)");
-      TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.05));
-      Console.WriteLine("Eps(0.1)");
-      TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.1));
-      Console.WriteLine("Eps(0.2)");
-      TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.2));
-      Console.WriteLine("Eps(0.5)");
-      TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.5));
+      TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.05));
+      //Console.WriteLine("Eps(0.1)");
+      //TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.1));
+      //Console.WriteLine("Eps(0.2)");
+      //TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.2));
+      //Console.WriteLine("Eps(0.5)");
+      //TestPolicyBernoulli(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.5));
     }
     [TestMethod]
@@ -40 +46 @@
       var seedForPolicy = globalRand.Next();
       var nArms = 10;
+      Console.WriteLine("Thompson (Gaussian)");
+      TestPolicyNormal(globalRand, nArms, new GaussianThompsonSamplingPolicy(new Random(seedForPolicy), nArms));
       Console.WriteLine("Random");
-      TestPolicyNormal(globalRand, nArms, new RandomPolicy(new Random(seedForPolicy), 10));
+      TestPolicyNormal(globalRand, nArms, new RandomPolicy(new Random(seedForPolicy), nArms));
       Console.WriteLine("UCB1");
-      TestPolicyNormal(globalRand, nArms, new UCB1Policy(10));
+      TestPolicyNormal(globalRand, nArms, new UCB1Policy(nArms));
       Console.WriteLine("UCB1Tuned");
-      TestPolicyNormal(globalRand, nArms, new UCB1TunedPolicy(10));
+      TestPolicyNormal(globalRand, nArms, new UCB1TunedPolicy(nArms));
       Console.WriteLine("UCB1Normal");
-      TestPolicyNormal(globalRand, nArms, new UCBNormalPolicy(10));
+      TestPolicyNormal(globalRand, nArms, new UCBNormalPolicy(nArms));
+      //Console.WriteLine("Exp3 (gamma=0.01)");
+      //TestPolicyNormal(globalRand, nArms, new Exp3Policy(new Random(seedForPolicy), nArms, 0.01));
+      //Console.WriteLine("Exp3 (gamma=0.05)");
+      //TestPolicyNormal(globalRand, nArms, new Exp3Policy(new Random(seedForPolicy), nArms, 0.05));
       Console.WriteLine("Eps(0.01)");
-      TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.01));
+      TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.01));
       Console.WriteLine("Eps(0.05)");
-      TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.05));
-      Console.WriteLine("Eps(0.1)");
-      TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.1));
-      Console.WriteLine("Eps(0.2)");
-      TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.2));
-      Console.WriteLine("Eps(0.5)");
-      TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), 10, 0.5));
+      TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.05));
+      //Console.WriteLine("Eps(0.1)");
+      //TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.1));
+      //Console.WriteLine("Eps(0.2)");
+      //TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.2));
+      //Console.WriteLine("Eps(0.5)");
+      //TestPolicyNormal(globalRand, nArms, new EpsGreedyPolicy(new Random(seedForPolicy), nArms, 0.5));
     }