public static double GetValue(State s, FoeQTable q)
{
var contxt = SolverContext.GetContext();
var model = contxt.CreateModel();
var actDecisions = new List <Decision>();
foreach (var action in Enum.GetNames(typeof(Action)))
{
var decision = new Decision(Domain.RealNonnegative, action);
model.AddDecision(decision);
actDecisions.Add(decision);
}
var val_Decis = new Decision(Domain.Real, "value");
model.AddDecision(val_Decis);
model.AddConstraint("probSumConst", actDecisions[0] + actDecisions[1] + actDecisions[2] + actDecisions[3] + actDecisions[4] == 1.0);
int cons_cnt = 0;
foreach (Action Player2 in Enum.GetValues(typeof(Action)))
{
var qConstVals = new List <double>();
foreach (Action currentPlayer in Enum.GetValues(typeof(Action)))
{
qConstVals.Add(q.getQval(s, currentPlayer, Player2));
}
model.AddConstraint("Const" + cons_cnt, qConstVals[0] * actDecisions[0] + qConstVals[1] * actDecisions[1] + qConstVals[2] * actDecisions[2] + qConstVals[3] * actDecisions[3] + qConstVals[4] * actDecisions[4] >= val_Decis);
++cons_cnt;
}
model.AddGoal("MaximizeV", GoalKind.Maximize, val_Decis);
contxt.Solve(new SimplexDirective());
contxt.ClearModel();
return(val_Decis.GetDouble());
}
private static void RunFoeQ()
{
var S = new StateSet();
var A = new JointActionSet();
var P = new Transition(S, A);
var Q_A = new FoeQTable();
var R = new Rewards();
var ERR = new Dictionary <int, double>();
var j = 0;
while (j <= 1000000)
{
//Initialize state according to Figure 4
var initialState = new State(2, 1, BallPossessor.B);
var done = false;
var currState = new State(3, 1, BallPossessor.B);
var actions = A.GetNextJointAction();
var playerAAction = actions.currplayer1Act;
var playerBAction = actions.player2act;
while (!done)
{
if (j % 20000 == 0)
{
Console.WriteLine(j);
}
var q_fig_4_initial = Q_A.getQval(initialState, Action.S, Action.X);
var nextState = P.GetNextState(currState, new JointAction(playerAAction, playerBAction));
var playerAReward = R.GetPlayerAReward(nextState);
var playerBReward = R.GetPlayerBReward(nextState);
if (playerAReward != 0.0 || playerBReward != 0.0)
{
done = true;
}
Q_A.UpdateQValue(currState, nextState, playerAAction, playerBAction, playerAReward);
if (currState.Equals(initialState) && playerAAction == Action.S && playerBAction == Action.X)
{
var q_fig_4 = Q_A.getQval(initialState, Action.S, Action.X);
var diff = Math.Abs(q_fig_4 - q_fig_4_initial);
ERR.Add(j, diff);
}
currState = nextState;
actions = A.GetNextJointAction();
playerAAction = actions.currplayer1Act;
playerBAction = actions.player2act;
++j;
}
}
using (StreamWriter sw = File.CreateText("output.csv"))
{
foreach (var kvp in ERR)
{
sw.WriteLine(kvp.Key + "," + kvp.Value);
}
}
}