Changeset 11977 for branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.GrammaticalOptimization/SequentialDecisionPolicies/GenericFunctionApproximationGrammarPolicy.cs

Timestamp:

02/11/15 03:01:59 (10 years ago)

Author:

gkronber

Message:

#2283 commit for 'realistic' (same settings for ant and symbreg) experiment

File:

• branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.GrammaticalOptimization/SequentialDecisionPolicies/GenericFunctionApproximationGrammarPolicy.cs (modified) (5 diffs)

branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.GrammaticalOptimization/SequentialDecisionPolicies/GenericFunctionApproximationGrammarPolicy.cs

@@ -62 +62 @@
 
 
-      /*
-      const double beta = 1;
+      
+      const double beta = 20;
       var w = from idx in Enumerable.Range(0, maxIdx)
               let afterStateQ = activeAfterStates[idx]
@@ -71 +71 @@
       selectedStateIdx = actionIndexMap[bestAction];
       Debug.Assert(selectedStateIdx >= 0);
-      */
-
-
+      
+
+      /*
       if (random.NextDouble() < 0.2) {
         selectedStateIdx = actionIndexMap[random.Next(maxIdx)];
@@ -91 +91 @@
         selectedStateIdx = actionIndexMap[bestIdxs[random.Next(bestIdxs.Count)]];
       }
-
-
-
-
+      */
       return true;
     }
@@ -132 +129 @@
 
     public double GetValue(string state) {
-      return problem.GetFeatures(state).Average(feature => GetWeight(feature));
+      return problem.GetFeatures(state).Sum(feature => GetWeight(feature));
     }
 
@@ -142 +139 @@
     private void UpdateWeights(string state, double reward) {
       double delta = reward - GetValue(state);
-      delta /= problem.GetFeatures(state).Count();
-      const double alpha = 0.001;
+      // delta /= problem.GetFeatures(state).Count();
+      //const double alpha = 0.01;
       foreach (var feature in problem.GetFeatures(state)) {
         featureTries[feature.Id] = GetFeatureTries(feature.Id) + 1;
         Debug.Assert(GetFeatureTries(feature.Id) >= 1);
-        //double alpha = 1.0 / GetFeatureTries(feature.Id);
-        //alpha = Math.Max(alpha, 0.01);
+        double alpha = 1.0 / GetFeatureTries(feature.Id);
+        alpha = Math.Max(alpha, 0.001);
 
         double w;