// Executes the given action
public void executeAction(Action a) {
// For each state variable, retrieve the DD for it, evaluate it
// w.r.t. the current state and build a new state.
ArrayList new_state = new ArrayList();
int c;
for (c = 0; c < (_nVars << 1); c++) {
new_state.add("-");
}
for (c = 0; c < (_nVars << 1); c++) {
Object cur_assign = _state.get(c);
if (cur_assign instanceof Boolean) {
// System.out.println(a._tmID2DD);
int nonprime_id = c + 1;
int prime_id = nonprime_id - _nVars;
Object DD = a._tmID2DD.get(new Integer(prime_id));
_state.set(prime_id - 1, TRUE);
double p_true = _mdp._context.evaluate(DD, _state);
// System.out.println("ID: " + nonprime_id + ", " + prime_id + ": " +
// _mdp._context.printNode(DD) + " -> " +
// _mdp._df.format(p_true));
new_state.set(c, (_r.nextFloat() < p_true) ? TRUE : FALSE);
}
}
_state = new_state;
}
// Derives QFunctions for the given value function and simulates the
// greedy policy for the given number of trials and steps per trial.
// Returns final value of every trial.
public ArrayList simulate(int trials, int steps, long rand_seed) {
ArrayList values = new ArrayList();
_r = new Random(rand_seed);
for (int trial = 1; trial <= trials; trial++) {
System.out.println("\n -----------\n Trial " + trial + "\n -----------");
// Initialize state
_state = new ArrayList();
_nVars = _mdp._alVars.size();
for (int c = 0; c < (_nVars << 1); c++) {
_state.add("-");
}
Iterator i = _mdp._alVars.iterator();
_vars = new TreeSet();
while (i.hasNext()) {
String s = (String) i.next();
if (!s.endsWith("\'")) {
Integer gid = (Integer) _mdp._tmVar2ID.get(s);
_vars.add(gid);
// Note: assign level (level is gid-1 b/c gids in order)
_state.set(gid.intValue() - 1, _r.nextBoolean() ? TRUE : FALSE);
}
}
// System.out.println(_mdp._context.printNode(_mdp._valueDD) + "\n" + _state);
double reward = _mdp._context.evaluate(_mdp._rewardDD, _state);
System.out.print(" " + PrintState(_state) + " " + MDP._df.format(reward));
// Run steps
for (int step = 1; step <= steps; step++) {
// Get action
Action a;
if (_bUseBasis) {
a = getBasisAction();
} else {
a = getAction();
}
// Execute action
executeAction(a);
// Update reward
reward =
(_mdp._bdDiscount.doubleValue() * reward)
+ _mdp._context.evaluate(_mdp._rewardDD, _state);
System.out.println(", a=" + a._sName);
System.out.print(
" " + PrintState(_state) + " " + MDP._df.format(reward) + ": " + "Step " + step);
}
values.add(new Double(reward));
System.out.println();
}
return values;
}
public static double Average(ArrayList l) {
double accum = 0.0d;
Iterator i = l.iterator();
while (i.hasNext()) {
Double d = (Double) i.next();
accum += d.doubleValue();
}
return (accum / (double) l.size());
}
// Fill in weights for specific counting aggregator approaches
//
// Basis function 1: Count of computers running
// Basis function 2: Count of computers running and connected to
// one other running computer
//
// Assuming action succeeds... compute next-state value based on action,
// choose best action.
public Action getBasisAction() {
int best_reboot = -1;
double best_reboot_val = -1d;
for (int c = 1; c <= _nVars; c++) {
ArrayList test_state = (ArrayList) _state.clone();
test_state.set(c - 1 + _nVars, TRUE);
double test_val = evalBasisState(test_state, W_BASIS_1, W_BASIS_2);
if (test_val > best_reboot_val) {
best_reboot_val = test_val;
best_reboot = c;
}
}
return (Action) _mdp._hmName2Action.get("reboot" + best_reboot);
}
public double evalBasisState(ArrayList state, double wb1, double wb2) {
int sum_basis_1 = 0;
int sum_basis_2 = 0;
// TODO: Code to get connection links... use current state to get status
// System.out.print(", Eval: " + state);
Action a = (Action) _mdp._hmName2Action.get("noreboot");
for (int c = 0; c < (_nVars << 1); c++) {
Object cur_assign = state.get(c);
if (cur_assign instanceof Boolean && ((Boolean) cur_assign).booleanValue()) {
sum_basis_1++;
// System.out.println(a._tmID2DD);
int nonprime_id = c + 1;
int prime_id = nonprime_id - _nVars;
Object DD = a._tmID2DD.get(new Integer(prime_id));
Set IDs = _mdp._context.getGIDs(DD);
// System.out.print(IDs + ": ");
Iterator it = IDs.iterator();
// System.out.print("[" + prime_id + " - ");
while (it.hasNext()) {
int c_id = ((Integer) it.next()).intValue() - 1; // nonprime array index
if (c_id < _nVars || c_id == c) { // Skip prime/np this var
continue;
}
// System.out.print(c_id + " ");
Object c_id_assign = state.get(c_id);
if (c_id_assign instanceof Boolean && ((Boolean) c_id_assign).booleanValue()) {
sum_basis_2++;
break;
}
}
// System.out.println("]");
}
}
double val = wb1 * sum_basis_1 + wb2 * sum_basis_2;
// System.out.println(state + ", b1: " + sum_basis_1 +
// ", b2: " + sum_basis_2 + ", total: " +
// _mdp._df.format(val));
return val;
}
public static String PrintList(ArrayList l) {
StringBuffer sb = new StringBuffer("[ ");
Iterator i = l.iterator();
while (i.hasNext()) {
Double d = (Double) i.next();
sb.append(MDP._df.format(d.doubleValue()) + " ");
}
sb.append("]");
return sb.toString();
}
public static String PrintState(ArrayList state) {
StringBuffer sb = new StringBuffer();
Iterator i = state.iterator();
while (i.hasNext()) {
Object o = i.next();
if (o instanceof Boolean) {
Boolean val = (Boolean) o;
sb.append((val.booleanValue() ? "." : "X"));
}
}
return sb.toString();
}
