Changeset 11742 for branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/BanditPolicies/GaussianThompsonSamplingPolicy.cs

Timestamp:

01/09/15 14:57:28 (9 years ago)

Author:

gkronber

Message:

#2283 refactoring

Location:

branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/BanditPolicies

Files:

• . (moved) (moved from branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/Policies)
• GaussianThompsonSamplingPolicy.cs (modified) (4 diffs)

branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/BanditPolicies/GaussianThompsonSamplingPolicy.cs

@@ -5 +5 @@
 using HeuristicLab.Common;
 
-namespace HeuristicLab.Algorithms.Bandits {
+namespace HeuristicLab.Algorithms.Bandits.BanditPolicies {
 
-  public class GaussianThompsonSamplingPolicy : IPolicy {
+  [Obsolete("Replaced by GenericThompsonSamplingPolicy(GaussianModel(0.5, 1.0, 0.1))")]
+  public class GaussianThompsonSamplingPolicy : IBanditPolicy {
     private bool compatibility;
 
@@ -22 +23 @@
     }
 
-    public int SelectAction(Random random, IEnumerable<IPolicyActionInfo> actionInfos) {
+    public int SelectAction(Random random, IEnumerable<IBanditPolicyActionInfo> actionInfos) {
       var myActionInfos = actionInfos.OfType<MeanAndVariancePolicyActionInfo>();
       int bestAction = -1;
@@ -38 +39 @@
         double theta;
         if (compatibility) {
+          // old code used for old experiments (preserved because it performed very well)
           if (tries < 2) return aIdx;
           var mu = sampleMean;
@@ -65 +67 @@
     }
 
-    public IPolicyActionInfo CreateActionInfo() {
+    public IBanditPolicyActionInfo CreateActionInfo() {
       return new MeanAndVariancePolicyActionInfo();
     }