Changeset 11792 for branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/GrammarPolicies

Timestamp:

01/16/15 18:26:35 (10 years ago)

Author:

gkronber

Message:

#2283 work-in-progress commit (does not compile)

File:

• branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/GrammarPolicies/GenericGrammarPolicy.cs (modified) (5 diffs)

branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/GrammarPolicies/GenericGrammarPolicy.cs

@@ -27 +27 @@
       out ReadonlySequence selectedState) {
       // only select states that are not yet done
-      afterStates = afterStates.Where(a => !done.Contains(CanonicalState(a.ToString()))).ToArray();
+      afterStates = afterStates.Where(a => !done.Contains(CanonicalState(a))).ToArray();
       if (!afterStates.Any()) {
         // fail because all follow states have already been visited => also disable the current state (if we can be sure that it has been fully explored)
-        throw new NotImplementedException();
-        //var curStateCanonical = CanonicalState(curState.ToString());
-        //if (curState.ToString().Length == curStateCanonical.Length)
-          done.Add(CanonicalState(curState.ToString()));
+
+        done.Add(CanonicalState(curState));
         selectedState = null;
         return false;
@@ -45 +43 @@
 
     private IBanditPolicyActionInfo GetStateInfo(ReadonlySequence state) {
-      var s = CanonicalState(state.ToString());
+      var s = CanonicalState(state);
       IBanditPolicyActionInfo info;
       if (!stateInfo.TryGetValue(s, out info)) {
@@ -57 +55 @@
       // the last state could be terminal
       var lastState = stateTrajectory.Last();
-      if (lastState.IsTerminal) done.Add(CanonicalState(lastState.ToString()));
+      if (lastState.IsTerminal) done.Add(CanonicalState(lastState));
 
       foreach (var state in stateTrajectory) {
@@ -70 +68 @@
 
     public int GetTries(ReadonlySequence state) {
-      var s = CanonicalState(state.ToString());
+      var s = CanonicalState(state);
       if (stateInfo.ContainsKey(s)) return stateInfo[s].Tries;
       else return 0;
@@ -76 +74 @@
 
     public double GetValue(ReadonlySequence state) {
-      var s = CanonicalState(state.ToString());
+      var s = CanonicalState(state);
       if (stateInfo.ContainsKey(s)) return stateInfo[s].Value;
       else return 0.0; // TODO: check alternatives
     }
 
-    protected string CanonicalState(string state) {
-      if (useCanonicalState) return problem.CanonicalRepresentation(state);
-      else return state;
+    protected string CanonicalState(ReadonlySequence state) {
+      if (useCanonicalState) {
+        if (state.IsTerminal)
+          return problem.CanonicalRepresentation(state.ToString());
+        else {
+          // for non-terminal phrases make sure we don't disable canonical states that have not yet been fully explored
+          // e.g. if for the ant problem we have the phrase lllS (and we are limited to 4 symbols) and lllr as well as llll are explored
+          // then we are not allowed to disable rS (canonical of lllS) because rS might not have been fully explored
+          // solution: we disable the state rS4 
+          return problem.CanonicalRepresentation(state.ToString()) + state.Length;
+        }
+      } else
+        return state.ToString();
     }
   }