Ejemplo n.º 1
0
 private GamblersWorldAction MaxAction(GamblersWorldState state)
 {
     return(_world
            .AvailableActions(state)
            .MaxBy(action => _policy.PAction(state, action))
            .First());
 }
Ejemplo n.º 2
0
        private static GamblersWorldAction FindBestAction(
            GamblersWorld world,
            GamblersWorldState state,
            GamblersValueTable valueTable,
            IGamblersWorldRewarder rewarder)
        {
            var maxActionValue = double.MinValue;
            var maxAction      = new GamblersWorldAction(0);

            foreach (var action in world.AvailableActions(state))
            {
                var actionValue = 0.0;

                foreach (var(nextState, pNextState) in world.PossibleStates(state, action))
                {
                    var nextStateValue = valueTable.Value(nextState);
                    var reward         = rewarder.Reward(state, nextState, action);

                    actionValue += pNextState * (reward + nextStateValue);
                }

                if (actionValue > maxActionValue)
                {
                    maxActionValue = actionValue;
                    maxAction      = action;
                }
            }

            return(maxAction);
        }
Ejemplo n.º 3
0
        private double CalculateValue(
            GamblersWorldState state, IGamblersPolicy policy, IGamblersWorldRewarder rewarder)
        {
            var newValue = 0.0;

            foreach (var action in _world.AvailableActions(state))
            {
                foreach (var(nextState, pNextState) in _world.PossibleStates(state, action))
                {
                    var reward = rewarder.Reward(state, nextState, action);
                    newValue +=
                        policy.PAction(state, action)
                        * pNextState
                        * (reward + Value(nextState));
                }
            }

            return(newValue);
        }