public Node CreateNode(IStateProtocol state)
{
var action = state.GetLastAction();
var playerIndex = state.GetLastPlayerIndex();
return(Node.CreateRootNode(playerIndex, action, state));
}
public void Update(Node node, IStateProtocol finalState)
{
var rewardDifferences = new double[MAX_PLAYER_COUNT];
for (uint i = 0; i < finalState.PlayerCount; i++)
{
var reward = finalState.GetReward(i);
var otherPlayerReward = double.MinValue;
for (uint j = 0; j < finalState.PlayerCount; j++)
{
if (i != j)
{
otherPlayerReward = Math.Max(otherPlayerReward, finalState.GetReward(j));
}
}
rewardDifferences[i] = Normalizer.Normalize(reward - otherPlayerReward);
}
var currentNode = node;
while (!currentNode.IsRoot)
{
var playerIndex = currentNode.PlayerIndex;
currentNode.AddReward(rewardDifferences[playerIndex]);
currentNode.IncreaseVisitCount();
currentNode = currentNode.Parent;
}
// Root node
currentNode.IncreaseVisitCount();
}
/// <summary>
/// Tính toán hành động tiếp theo cho State này
/// </summary>
public void ComputeMove(IStateProtocol state)
{
if (IsComputing)
{
return;
}
_beginPoint = DateTime.Now.Millisecond;
Func <mcts.IActionProtocol> act = () => {
return(_callback?.Invoke(state));
};
_future = new Task <mcts.IActionProtocol>(act);
_future.Start();
}
public void Update(Node node, IStateProtocol finalState)
{
var rewards = new double[MAX_PLAYER_COUNT];
for (uint i = 0; i < finalState.PlayerCount; i++)
{
rewards[i] = Normalizer.Normalize(finalState.GetReward(i));
}
var currentNode = node;
while (!currentNode.IsRoot)
{
var playerIndex = currentNode.PlayerIndex;
currentNode.AddReward(rewards[playerIndex]);
currentNode.IncreaseVisitCount();
currentNode = currentNode.Parent;
}
// Root node
currentNode.IncreaseVisitCount();
}
public IActionProtocol ComputeAction(IStateProtocol state, Action <string> logger)
{
// FIXME: chưa làm
return(null);
}
public StateTree(IStateProtocol state)
{
_root = new StateLeaf(state);
_currentLeaf = _root;
}
private void BackPropagation(Node expandedNode, IStateProtocol finalState)
{
_backPropagationStrategy.Update(expandedNode, finalState);
}
private Node(NodeBase parent, uint playerIndex, IActionProtocol action, IStateProtocol state) : base(parent, playerIndex)
{
Action = action;
State = state;
}
public static Node CreateNode(Node parent, uint playerIndex, IActionProtocol action, IStateProtocol state)
{
return(new Node(parent, playerIndex, action, state));
}
public static Node CreateRootNode(uint playerIndex, IActionProtocol action, IStateProtocol state)
{
return(new Node(null, playerIndex, action, state));
}