private List <HashSet <TState> > GetGreedyDependencies(IDeterministicPolicy <TState> policy, TState state)
{
var list = new List <HashSet <TState> >();
var iterations = ValueFunctions[policy].Count - 1;
var valueFunction = ValueFunctions[policy][iterations];
var allowedActions = policy.GetAllowedActions(state).Where(a => a != null);
var prevPolicy = Policies.Find(policy)?.Previous?.Value;
var nextStates = new HashSet <TState>();
foreach (var action in allowedActions)
{
var actionStates = action[state].Where(s => !valueFunction.HasState(s));
nextStates.UnionWith(actionStates);
}
list.Add(nextStates);
while (iterations-- > 0)
{
var set = new HashSet <TState>();
valueFunction = ValueFunctions[policy][iterations];
foreach (var nextState in nextStates)
{
if (!UseReachableStateSpace && !policy.HasOptimal(nextState) && prevPolicy != null)
{
policy[nextState] = GetOptimalActionEx(prevPolicy, nextState);
}
var action = policy[nextState];
if (action == null)
{
continue;
}
// IEnumerable<TState> actionStates;
// lock (action)
var actionStates = action[nextState].Where(s => !valueFunction.HasState(s));
set.UnionWith(actionStates);
}
list.Add(set);
nextStates = set;
}
return(list);
}
