Created
February 10, 2017 20:03
-
-
Save caffeine-potent/5111c57d475d8091f1433790fcb5fb3e to your computer and use it in GitHub Desktop.
Epsilon Greedy from "Bandit Algorithms"
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
class EpsGreedy() | |
def __init__(self, number_of_bandits, epsilon, start_greedy= True): | |
self.count = np.zeros(number_of_bandits) | |
self.scores = np.array([int(!start_greedy)] * number of bandits) | |
self.epsilon = epsilon | |
self.bandit_count = number_of_bandits | |
def select_arm(): | |
choice = np.random.binomial(1,self.epsilon): | |
if choice = 1: #EXPLORE | |
return np.random.randint(self.number_of_bandits) | |
else: | |
return self.scores.argmax() | |
def update(self, bandit_index, reward): | |
self.count[bandit_index] = self.count[bandit_index] + 1 | |
n = self.count[bandit_index] | |
score = self.scores[bandit_index] | |
new_value = (n-1/float(n)) * score + (1/float(n)) * reward | |
self.score[bandit_index] = new_value | |
class Binomial_Bandit: | |
def __init__(self, p_of_payoff): | |
self.probability = p_of_payoff | |
def get_reward(): | |
return np.binomial(1,self.probability) | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment