增强学习--Q-leraning

Q-learning

实例代码

 import numpy as np
 import random
 from environment import Env
 from collections import defaultdict
 
 class QLearningAgent:
     def __init__(self, actions):
         # actions = [0, 1, 2, 3]
         self.actions = actions
         self.learning_rate = 0.01
         self.discount_factor = 0.9
         self.epsilon = 0.1
         self.q_table = defaultdict(lambda: [0.0, 0.0, 0.0, 0.0])#待更新q表
 
     # update q function with sample <s, a, r, s'>
     def learn(self, state, action, reward, next_state):
         current_q = self.q_table[state][action]
         # using Bellman Optimality Equation to update q function
         new_q = reward + self.discount_factor * max(self.q_table[next_state])
         self.q_table[state][action] += self.learning_rate * (new_q - current_q)#更新公式,off-policy
 
     # get action for the state according to the q function table
     # agent pick action of epsilon-greedy policy
     def get_action(self, state):
         #epsilon-greedy policy
         if np.random.rand() < self.epsilon:
             # take random action
             action = np.random.choice(self.actions)
         else:
             # take action according to the q function table
             state_action = self.q_table[state]
             action = self.arg_max(state_action)
         return action
 
     @staticmethod
     def arg_max(state_action):
         max_index_list = []
         max_value = state_action[0]
         for index, value in enumerate(state_action):
             if value > max_value:
                 max_index_list.clear()
                 max_value = value
                 max_index_list.append(index)
             elif value == max_value:
                 max_index_list.append(index)
         return random.choice(max_index_list)
 
 if __name__ == "__main__":
     env = Env()
     agent = QLearningAgent(actions=list(range(env.n_actions)))
 
     for episode in range(1000):
         state = env.reset()
 
         while True:
             env.render()
 
             # take action and proceed one step in the environment
             action = agent.get_action(str(state))
             next_state, reward, done = env.step(action)
 
             # with sample <s,a,r,s'>, agent learns new q function
             agent.learn(str(state), action, reward, str(next_state))
 
             state = next_state
             env.print_value_all(agent.q_table)
 
             # if episode ends, then break
             if done:
                 break