source: branches/HeuristicLab.Problems.GrammaticalOptimization/Main/Program.cs @ 11732

using System;
using System.Collections.Generic;
using System.Data;
using System.Diagnostics;
using System.Globalization;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using HeuristicLab.Algorithms.Bandits;
using HeuristicLab.Algorithms.Bandits.Models;
using HeuristicLab.Algorithms.GrammaticalOptimization;
using HeuristicLab.Problems.GrammaticalOptimization;
using HeuristicLab.Problems.GrammaticalOptimization.SymbReg;

namespace Main {
  class Program {
    static void Main(string[] args) {
      CultureInfo.DefaultThreadCurrentCulture = CultureInfo.InvariantCulture;

      RunDemo();
      //RunGridTest();
    }

    private static void RunGridTest() {
      int maxIterations = 200000; // for poly-10 with 50000 evaluations no successful try with hl yet
      //var globalRandom = new Random(31415);
      var localRandSeed = 31415;
      var reps = 20;

      var policies = new Func<IPolicy>[]
        {
         () => new GaussianThompsonSamplingPolicy(), 
         () => new GaussianThompsonSamplingPolicy(true), 
         () => new GenericThompsonSamplingPolicy(new GaussianModel(0.5, 1)), 
         () => new BernoulliThompsonSamplingPolicy(),
         () => new GenericThompsonSamplingPolicy(new BernoulliModel(1, 1)), 
         () => new RandomPolicy(), 
         () => new EpsGreedyPolicy(0.01), 
         () => new EpsGreedyPolicy(0.05), 
         () => new EpsGreedyPolicy(0.1), 
         () => new EpsGreedyPolicy(0.2), 
         () => new EpsGreedyPolicy(0.5), 
         () => new UCTPolicy(0.1),
         () => new UCTPolicy(0.5),
         () => new UCTPolicy(1),
         () => new UCTPolicy(2),
         () => new UCTPolicy( 5),
         () => new UCTPolicy( 10),
         () => new UCB1Policy(), 
         () => new UCB1TunedPolicy(), 
         () => new UCBNormalPolicy(), 
         () => new BoltzmannExplorationPolicy(0.1), 
         () => new BoltzmannExplorationPolicy(0.5),
         () => new BoltzmannExplorationPolicy(1),
         () => new BoltzmannExplorationPolicy(5),
         () => new BoltzmannExplorationPolicy(10),
         () => new BoltzmannExplorationPolicy(20),
         () => new BoltzmannExplorationPolicy(100),
         () => new ChernoffIntervalEstimationPolicy( 0.01), 
         () => new ChernoffIntervalEstimationPolicy( 0.05),
         () => new ChernoffIntervalEstimationPolicy( 0.1),
         () => new ChernoffIntervalEstimationPolicy( 0.2),
         // (rand) => new ThresholdAscentPolicy(10, 0.01), 
         // (rand) => new ThresholdAscentPolicy(10, 0.05),
         // (rand) => new ThresholdAscentPolicy(10, 0.1),
         // (rand) => new ThresholdAscentPolicy(10, 0.2),
         // (rand) => new ThresholdAscentPolicy(100, 0.01), 
         // (rand) => new ThresholdAscentPolicy(100, 0.05),
         // (rand) => new ThresholdAscentPolicy(100, 0.1),
         // (rand) => new ThresholdAscentPolicy(100, 0.2),
         // (rand) => new ThresholdAscentPolicy(1000, 0.01), 
         // (rand) => new ThresholdAscentPolicy(1000, 0.05),
         // (rand) => new ThresholdAscentPolicy(1000, 0.1),
         // (rand) => new ThresholdAscentPolicy(1000, 0.2),
         // (rand) => new ThresholdAscentPolicy(5000, 0.01), 
         // (rand) => new ThresholdAscentPolicy(10000, 0.01), 
        };

      foreach (var problem in new Tuple<IProblem, int>[]
        {
          Tuple.Create((IProblem)new SantaFeAntProblem(), 17), 
          Tuple.Create((IProblem)new SymbolicRegressionPoly10Problem(), 23), 
        })
        foreach (var randomTries in new int[] { 1, 10, /* 5, 100 /*, 500, 1000 */}) {
          foreach (var policy in policies) {
            var myRandomTries = randomTries;
            var localRand = new Random(localRandSeed);
            var options = new ParallelOptions();
            options.MaxDegreeOfParallelism = 1;
            Parallel.For(0, reps, options, (i) => {
              //var t = Task.Run(() => {
              Random myLocalRand;
              lock (localRand)
                myLocalRand = new Random(localRand.Next());

              //for (int i = 0; i < reps; i++) {

              int iterations = 0;
              var globalStatistics = new SentenceSetStatistics();

              // var problem = new SymbolicRegressionPoly10Problem();
              // var problem = new SantaFeAntProblem();
              //var problem = new PalindromeProblem();
              //var problem = new HardPalindromeProblem();
              //var problem = new RoyalPairProblem();
              //var problem = new EvenParityProblem();
              var alg = new MctsSampler(problem.Item1, problem.Item2, myLocalRand, myRandomTries, policy()); // TODO: Make sure we generate the same random numbers for each experiment
              //var alg = new ExhaustiveBreadthFirstSearch(problem, 25);
              //var alg = new AlternativesContextSampler(problem, 25);

              alg.SolutionEvaluated += (sentence, quality) => {
                iterations++;
                globalStatistics.AddSentence(sentence, quality);
                if (iterations % 10000 == 0) {
                  Console.WriteLine("{0,4} {1,7} {2,5} {3,25} {4}", alg.treeDepth, alg.treeSize, myRandomTries, policy(), globalStatistics);
                }
              };


              alg.Run(maxIterations);

              //Console.WriteLine("{0,5} {1} {2}", randomTries, policyFactory(1), globalStatistics);
              //}
              //});
              //tasks.Add(t);
            });
          }
        }
      //Task.WaitAll(tasks.ToArray());
    }

    private static void RunDemo() {
      // TODO: test with eps-greedy using max instead of average as value (seems to work well for symb-reg! explore further!)
      // TODO: implement GaussianWithUnknownMeanAndVariance Model for Thompson Sampling (verify with unit test if correct mean and variance is identified)
      // TODO: separate value function from policy
      // TODO: debug and verify implementation variants of Gaussian Thompson Sampling with unit test
      // TODO: refactor Policies to use banditInfos (policies are factories for bandit infos and bandit info only has an update routine, each policy works only with it's type of banditinfo)
      // TODO: in contextual MCTS store a bandit info for each node in the _graph_ and also update all bandit infos of all parents
      // TODO: exhaustive search with priority list
      // TODO: warum funktioniert die alte Implementierung von GaussianThompson besser für SantaFe als alte? Siehe Vergleich: alte vs. neue implementierung GaussianThompsonSampling
      // TODO: why does GaussianThompsonSampling work so well with MCTS for the artificial ant problem?
      // TODO: wie kann ich sampler noch vergleichen bzw. was kann man messen um die qualität des samplers abzuschätzen (bis auf qualität und iterationen bis zur besten lösung) => ziel schnellere iterationen zu gutem ergebnis
      // TODO: likelihood für R=1 bei Gaussian oder GaussianMixture einfach berechenbar?
      // TODO: research thompson sampling for max bandit?
      // TODO: ausführlicher test von strategien für k-armed max bandit
      // TODO: verify TA implementation using example from the original paper      
      // TODO: compare results for different policies also for the symb-reg problem
      // TODO: separate policy from MCTS tree data structure to allow sharing of information over disconnected parts of the tree (semantic equivalence)
      // TODO: implement thompson sampling for gaussian mixture models
      // TODO: implement inspection for MCTS (eventuell interactive command line für statistiken aus dem baum anzeigen)
      // TODO: implement ACO-style bandit policy
      // TODO: implement sequences that can be manipulated in-place (instead of strings), alternatives are also stored as sequences, for a sequence the index of the first NT-symb can be stored 
      // TODO: gleichzeitige modellierung von transformierter zielvariable (y, 1/y, log(y), exp(y), sqrt(y), ...)
      // TODO: vergleich bei complete-randomly möglichst kurze sätze generieren vs. einfach zufällig alternativen wählen
      // TODO: reward discounting (für veränderliche reward distributions über zeit). speziellen unit-test dafür erstellen
      // TODO: constant optimization 


      int maxIterations = 100000;
      int iterations = 0;
      var sw = new Stopwatch();
      double bestQuality = 0;
      string bestSentence = "";
      var globalStatistics = new SentenceSetStatistics();
      var random = new Random();

      var problem = new SymbolicRegressionPoly10Problem();
      //var problem = new SantaFeAntProblem(); // good results e.g. with       var alg = new MctsSampler(problem, 17, random, 1, (rand, numActions) => new ThresholdAscentPolicy(numActions, 500, 0.01));
      //var problem = new SymbolicRegressionProblem("Tower"); // very good results e.g. new EpsGreedyPolicy(0.2) using max reward as quality !!!
      //var problem = new PalindromeProblem();
      //var problem = new HardPalindromeProblem();
      //var problem = new RoyalPairProblem();
      //var problem = new EvenParityProblem();
      var alg = new MctsSampler(problem, 23, random, 10, new EpsGreedyPolicy(0.2)); // GaussianModelWithUnknownVariance (and Q= 0.99-quantil) works well for Ant
      //var alg = new ExhaustiveBreadthFirstSearch(problem, 17);
      //var alg = new AlternativesContextSampler(problem, random, 17, 4, (rand, numActions) => new RandomPolicy(rand, numActions));
      //var alg = new ExhaustiveDepthFirstSearch(problem, 17);
      // var alg = new AlternativesSampler(problem, 17);
      // var alg = new RandomSearch(problem, random, 17);
      // var alg = new ExhaustiveRandomFirstSearch(problem, random, 17);

      alg.FoundNewBestSolution += (sentence, quality) => {
        bestQuality = quality;
        bestSentence = sentence;
        Console.WriteLine("{0,4} {1,7} {2}", alg.treeDepth, alg.treeSize, globalStatistics);
      };
      alg.SolutionEvaluated += (sentence, quality) => {
        iterations++;
        globalStatistics.AddSentence(sentence, quality);
        if (iterations % 1000 == 0) {
          alg.PrintStats();
        }
        if (iterations % 10000 == 0) {
          //Console.WriteLine("{0,10} {1,10:F5} {2,10:F5} {3}", iterations, bestQuality, quality, sentence);
          //Console.WriteLine("{0,4} {1,7} {2}", alg.treeDepth, alg.treeSize, globalStatistics);
          Console.WriteLine("{0,4} {1,7} {2}", alg.treeDepth, alg.treeSize, globalStatistics);
        }
      };


      sw.Start();

      alg.Run(maxIterations);

      sw.Stop();

      Console.WriteLine("{0,10} Best soultion: {1,10:F5} {2}", iterations, bestQuality, bestSentence);
      Console.WriteLine("{0:F2} sec {1,10:F1} sols/sec {2,10:F1} ns/sol",
        sw.Elapsed.TotalSeconds,
        maxIterations / (double)sw.Elapsed.TotalSeconds,
        (double)sw.ElapsedMilliseconds * 1000 / maxIterations);
    }
  }
}