/*
* Uses the UCB for Trees algorithm to select children for exploration.
* If any children are unexplored, a child is selected from them uniformly
* at random. Otherwise, the child with the maximal UCB is chosen.
*/
private ChanceNode SelectChildToVisit(DecisionNode node)
{
double maxUCB = double.MinValue;
ChanceNode bestChild = null;
var unvisitedChildren = new List <ChanceNode>();
foreach (var child in node.Children.Values)
{
if (child.VisitCount == 0)
{
unvisitedChildren.Add(child);
}
else
{
double ucb = CalculateUCB(node, child);
if (ucb > maxUCB)
{
maxUCB = ucb;
bestChild = child;
}
}
}
if (unvisitedChildren.Count == 0)
{
return(bestChild);
}
else
{
return(RandomProvider.Select(unvisitedChildren));
}
}
private double CalculateUCB(DecisionNode parent, ChanceNode child)
{
double v = child.AverageValue;
double n = Math.Log(parent.VisitCount) / child.VisitCount;
return(v + EXPLORATION_RATE * Math.Sqrt(n));
}
private double GetChanceNodeValue(ChanceNode node)
{
var child = node.GenerateChild();
return(SampleSearchTree(child));
}