Rev | Line | |
---|
[11710] | 1 | using System;
|
---|
| 2 | using System.Collections.Generic;
|
---|
| 3 | using System.Linq;
|
---|
| 4 | using System.Text;
|
---|
| 5 | using System.Threading.Tasks;
|
---|
| 6 |
|
---|
| 7 | namespace HeuristicLab.Algorithms.Bandits {
|
---|
| 8 | public class UCB1Policy : BanditPolicy {
|
---|
| 9 | private readonly int[] tries;
|
---|
| 10 | private readonly double[] sumReward;
|
---|
| 11 | private int totalTries = 0;
|
---|
| 12 | public UCB1Policy(int numActions)
|
---|
| 13 | : base(numActions) {
|
---|
| 14 | this.tries = new int[NumActions];
|
---|
| 15 | this.sumReward = new double[NumActions];
|
---|
| 16 | }
|
---|
| 17 |
|
---|
| 18 | public override int SelectAction() {
|
---|
| 19 | int bestAction = -1;
|
---|
| 20 | double bestQ = double.NegativeInfinity;
|
---|
| 21 | for (int i = 0; i < NumActions; i++) {
|
---|
| 22 | if (tries[i] == 0) return i;
|
---|
| 23 | var q = sumReward[i] / tries[i] + Math.Sqrt((2 * Math.Log(totalTries)) / tries[i]);
|
---|
| 24 | if (q > bestQ) {
|
---|
| 25 | bestQ = q;
|
---|
| 26 | bestAction = i;
|
---|
| 27 | }
|
---|
| 28 | }
|
---|
| 29 | return bestAction;
|
---|
| 30 | }
|
---|
| 31 | public override void UpdateReward(int action, double reward) {
|
---|
| 32 | totalTries++;
|
---|
| 33 | tries[action]++;
|
---|
| 34 | sumReward[action] += reward;
|
---|
| 35 | }
|
---|
| 36 | public override void Reset() {
|
---|
| 37 | totalTries = 0;
|
---|
| 38 | Array.Clear(tries, 0, tries.Length);
|
---|
| 39 | Array.Clear(sumReward, 0, sumReward.Length);
|
---|
| 40 | }
|
---|
| 41 | }
|
---|
| 42 | }
|
---|
Note: See
TracBrowser
for help on using the repository browser.