Greedy.txt

//By: Annahita Doulatshahi
//500608597

import java.util.concurrent.ThreadLocalRandom;

public class Greedy implements LearnAgent{
	static private int probBestChoice = 70;
	//epsilon = 100-probBestChoice;
	private int randomNum, index;
	private double estimateReward;
	private double[] actionList;
	private int[] actionCount;
	
	public Greedy(){
		
		estimateReward = 0;
		actionList = new double[10];
		actionCount = new int[10];
	}
	
	public int chooseAction(){	//pick a Bandit from 1-10
		randomNum = ThreadLocalRandom.current().nextInt(1,101);
		index = ThreadLocalRandom.current().nextInt(0,10);
		
		if (probBestChoice >= randomNum){
			double max = 0;
			for (int i = 0;i < 10;i++){
				if (max < actionList[i]){
					max=actionList[i];
					index=i;
				}
			}
		}
		
		actionCount[index]+=1;
		return index;
	}
	
	public void processReward(int reward){
		estimateReward = actionList[index]; 
		actionList[index] = estimateReward + ((1/actionCount[index])*(reward-estimateReward));
	}
	
}