devbox
笔触狂放9
这个屌丝很懒，什么也没留下！
关注作者
热门标签
热门文章
当前位置:   article > 正文

【reinforcement learning】Deep Q-Learning(DQN)简介

作者：笔触狂放9 | 2024-03-23 14:53:00

赞

踩

deep q-learning

Deep Q-Learning(DQN)

一、什么是DQN

（一）为什么出现了DQN？

    在Q-Learning和Sarsa算法中，我们使用了一种数据结构：Q表，用Q表存储所有的状态以及每个action的Q值。在现实问题中，不仅action种类可能非常多，state的数量更可能是指数级，这就为我们存储Q表和查找对应状态带来了很大的麻烦。在机器学习中，我们可以将state和action输入到神经网络中，神经网络分析后输出action的Q值，然后我们再根据Q-learning的原则执行之后的操作。
    通过神经网络，我们省去了存储和查找的时间，同时网络还能够捕捉到一些细节特征，极大的促进了强化学习的发展。

（二）网络是如何更新的？

    在Q-Learning中，我们更新参数时，需要两个值：Q现实和Q估计，这两个值都记录在Q表中。
    在DQN中，我们要更新神经网络的参数，同样需要这两个值，那如何获得呢？
    首先我们通过神经网络预测出Q估计：
Q ( s , a 1 )     Q ( s , a 2 ) Q(s,a1)\ \ \ Q(s,a2) Q(s,a1)   Q(s,a2)
    Q现实也是神经网络预测出的Q值，不过是对s的下一步s’的估计，一般由target network（后面会提到）给出：
Q 现 实 = R + γ ∗ max ⁡ [ Q ( s ′ , a 1 ) , Q ( s ′ , s 2 ) ] Q_{现实}=R+\gamma*\max[Q(s',a1),Q(s',s2)] Q现实​=R+γ∗max[Q(s′,a1),Q(s′,s2)]
    更新参数：
α ( Q 现 实 − Q 估 计 ) \alpha(Q_{现实}-Q_{估计}) α(Q现实​−Q估计​)

（三）两大法宝为DQN锦上添花

  1. Experience replay
    为DQN建立一个记忆库，将过去或者别人的一些学习经验存在记忆库中，在每次更新神经网络时，可以抽取一些经历进行学习，这样打乱了经历之间的相关性, 也使得神经网络更新更有效率。
  2. Fixed Q-targets
    同样是用来打乱了经历之间的相关性。我们在DQN中建立两个结构相同参数不同的神经网络。预测 Q 估计 的神经网络具备最新的参数, 而预测 Q 现实 的神经网络使用的参数则是很久以前的参数。

二、DQN的算法过程

Initialize replay memort D D D to capacity N N N
Initialize action-value function Q Q Q with randon weights θ \theta θ
Initialize target action-value function Q ^ \hat{Q} Q^​ with weight θ − = θ \theta^-=\theta θ−=θ
For episode = 1,M do:
  Initialize sequence s 1 = { x 1 } s_1=\{x_1\} s1​={x1​} and preprocessed sequence ϕ = ϕ ( s 1 ) \phi=\phi(s_1) ϕ=ϕ(s1​)
  For t=1,T *do
    With probablity ε \varepsilon ε select a random action a t a_t at​
    otherwise select a t = arg ⁡   max ⁡ a Q ( ϕ ( s t ) , a : θ ) a_t=\arg\ \max_aQ(\phi(s_t),a:\theta) at​=arg maxa​Q(ϕ(st​),a:θ)
    Execute action a t a_t at​ in emulator and observe reward r t r_t rt​ and image x t + 1 x_{t+1} xt+1​
    Set s t + 1 = s t , a t , x t + 1 s_{t+1}=s_t,a_t,x_{t+1} st+1​=st​,at​,xt+1​ and preprocess ϕ t + ! = ϕ ( s t + 1 ) \phi_{t+!}=\phi(s_{t+1}) ϕt+!​=ϕ(st+1​)
    Store transition ( ϕ t , a t , r t , ϕ t + 1 ) (\phi_t,a_t,r_t,\phi_{t+1}) (ϕt​,at​,rt​,ϕt+1​) in D D D
    Sample random minibatch of transitions ( ϕ j , a j , r j , ϕ j + 1 ) (\phi_j,a_j,r_j,\phi_{j+1}) (ϕj​,aj​,rj​,ϕj+1​) from D D D
    Set y j = { r j i f   e p i s o d e   t e r m i m a t e s   a t   s t e p   j + 1 r j + γ max ⁡ a ′ Q ^ ( ϕ j + 1 , a ′ : θ − ) o t h e r w i z e y_j=\left\{

rjif episode termimates at step j+1rj+γmaxa′Q^(ϕj+1,a′:θ−)otherwize
\right. yj​=⎩⎨⎧​​rj​rj​+γa′max​Q^​(ϕj+1​,a′:θ−)​if episode termimates at step j+1otherwize​
    Permorm a gradient descent step on ( y j − Q ( ϕ j , a j : θ ) ) 2 (y_j-Q(\phi_j,a_j:\theta))^2 (yj​−Q(ϕj​,aj​:θ))2 with respect to the network parameters θ \theta θ
    Every C C C steps reset Q ^ = Q \hat{Q}=Q Q^​=Q
   End For
End For

（三）实现代码

详细信息可查看 莫烦 强化学习

环境代码详见 maze_env.py
【reinforcement learning】Q-Learning简介

main.py

from maze_env import Maze
from RL_brain import DeepQNetwork

def run_maze():
    step = 0
    for episode in range(600):
        observation = env.reset()
        while True:
            env.render()
            action = RL.choose_action(observation)
            observation_,reward,done = env.step(action)
            RL.store_transition(observation,action,reward,observation_)
            if(step > 200) and (step % 5 ==0):
                RL.learn()
            observation = observation_
            if done:
                break
            step += 1
    print('game over')
    env.destroy()


if __name__ == "__main__":
    env = Maze()
    RL = DeepQNetwork(env.n_actions,env.n_features,learning_rate=0.01,
                      reward_decay=0.9,e_greedy=0.9,replace_target_iter=200,
                      memory_size=2000)
    env.after(100,run_maze)
    env.mainloop()
    RL.plot_cost()


  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31

RL_brain.py


import numpy as np
import pandas as pd
import tensorflow as tf

np.random.seed(1)
tf.set_random_seed(1)


# Deep Q Network off-policy
class DeepQNetwork:

    def __init__(
            self,
            n_actions,
            n_features,
            learning_rate=0.01,
            reward_decay=0.9,
            e_greedy=0.9,
            replace_target_iter=300,
            memory_size=500,
            batch_size=32,
            e_greedy_increment=None,
            output_graph=False,
    ):
        self.n_actions = n_actions
        self.n_features = n_features
        self.lr = learning_rate
        self.gamma = reward_decay
        self.epsilon_max = e_greedy
        self.replace_target_iter = replace_target_iter
        self.memory_size = memory_size
        self.batch_size = batch_size
        self.epsilon_increment = e_greedy_increment
        self.epsilon = 0 if e_greedy_increment is not None else self.epsilon_max

        self.learn_step_counter = 0

        self.memory =np.zeros((self.memory_size,n_features*2+2))

        self._build_net()
        t_params = tf.get_collection('target_net_params')
        e_params = tf.get_collection('eval_net_params')
        self.replace_target_op = [tf.assign(t, e) for t, e in zip(t_params, e_params)]
        self.sess = tf.Session()

        if output_graph:
            tf.summary.FileWriter("./logs/",self.sess.graph)

        self.sess.run(tf.global_variables_initializer())
        self.cost_his = []


    def _build_net(self):
        # enval_network
        self.s = tf.placeholder(tf.float32,[None,self.n_features],name='s')
        self.q_target = tf.placeholder(tf.float32,[None,self.n_actions],name='Q_target')

        with tf.variable_scope('eval_net'):
            c_names,n_l1,w_initializer,b_initializer = ['eval_net_params',tf.GraphKeys.GLOBAL_VARIABLES],10,tf.random_normal_initializer(0.,0.3),tf.constant_initializer(0.1)

            with tf.variable_scope('l1'):
                w1 = tf.get_variable('w1',[self.n_features,n_l1],initializer=w_initializer,collections=c_names)
                b1 = tf.get_variable('b1',[1,n_l1],initializer=b_initializer,collections=c_names)
                l1 = tf.nn.relu(tf.matmul(self.s,w1)+b1)

            with tf.variable_scope('l2'):
                w2 = tf.get_variable('w2',[n_l1,self.n_actions],initializer=w_initializer,collections=c_names)
                b2 = tf.get_variable('b2',[1,self.n_actions],initializer=b_initializer,collections=c_names)
                self.q_eval = tf.matmul(l1,w2)+b2

        with tf.variable_scope('loss'):
            self.loss = tf.reduce_mean(tf.squared_difference(self.q_target,self.q_eval))
        with tf.variable_scope('train'):
            self._train_op = tf.train.RMSPropOptimizer(self.lr).minimize(self.loss)


        #target_network
        self.s_ = tf.placeholder(tf.float32,[None,self.n_features],name='s_')
        with tf.variable_scope('target_net'):
            c_names = ['target_net_params',tf.GraphKeys.GLOBAL_VARIABLES]
            with tf.variable_scope('l1'):
                w1 = tf.get_variable('W1',[self.n_features,n_l1],initializer=w_initializer,collections=c_names)
                b1 = tf.get_variable('b1',[1,n_l1],initializer=b_initializer,collections=c_names)
                l1 = tf.nn.relu(tf.matmul(self.s_,w1)+b1)

            with tf.variable_scope('l2'):
                w2 = tf.get_variable('w2',[n_l1,self.n_actions],initializer=w_initializer,collections=c_names)
                b2 = tf.get_variable('b2',[1,self.n_actions],initializer=b_initializer,collections=c_names)
                self.q_next = tf.matmul(l1,w2)+b2


    def store_transition(self,s,a,r,s_):
        if not hasattr(self,'memory_counter'):
            self.memory_counter = 0

        transition = np.hstack((s,[a,r],s_))

        index = self.memory_counter % self.memory_size
        self.memory[index,:] = transition

        self.memory_counter += 1
#
#
    def choose_action(self,observation):
        observation = observation[np.newaxis, :]
        if np.random.uniform() < self.epsilon:
            actions_value = self.sess.run(self.q_eval , feed_dict={self.s : observation})
            action = np.argmax(actions_value)

        else:
            action = np.random.randint(0,self.n_actions)

        return action

    def _replace_target_params(self):
        t_params = tf.get_collection('target_net_params')
        e_params = tf.get_collection('eval_net_params')
        self.sess.run([tf.assign(t,e) for t,e in zip(t_params,e_params)])

    def learn(self):
        if self.learn_step_counter % self.replace_target_iter == 0:
            self._replace_target_params()
            print('\nreplace......\n')

        if self.memory_counter > self.memory_size:
            sample_index = np.random.choice(self.memory_size,size=self.memory_size)
        else:
            sample_index = np.random.choice(self.memory_counter,self.batch_size)

        batch_memory = self.memory[sample_index,:]

        q_next,q_eval = self.sess.run(
            [self.q_next,self.q_eval],
            feed_dict={
                self.s_ : batch_memory[:,-self.n_features],
                self.s : batch_memory[:,self.n_features]
            }
        )

        q_target = q_eval.copy()
        batch_index = np.arange(self.batch_size,dtype=np.int32)
        eval_act_index = batch_memory[:,self.n_features].astype(int)
        reward = batch_memory[:,self.n_features+1]

        q_target[batch_index,eval_act_index] = reward + self.gamma*np.max(q_next,axis=1)

        _, self.cost = self.sess.run([self._train_op,self.loss],
                                    feed_dict={
                                        self.s : batch_memory.ioc[:,:self.n_features],
                                        self.q_target:q_target
                                    })
        self.cost_his.append(self.cost)

        self.epsilon = self.epsilon + self.epsilon_increment if self.epsilon < self.epsilon_max else self.epsilon_max
        self.learn_step_counter += 1

    def plot_cost(self):
        import matplotlib.pyplot as plt
        plt.plot(np.arange(len(self.cost_his)), self.cost_his)
        plt.ylabel('Cost')
        plt.xlabel('training steps')
        plt.show()




    def choose_action(self, observation):
        # to have batch dimension when feed into tf placeholder
        observation = observation[np.newaxis, :]

        if np.random.uniform() < self.epsilon:
            # forward feed the observation and get q value for every actions
            actions_value = self.sess.run(self.q_eval, feed_dict={self.s: observation})
            action = np.argmax(actions_value)
        else:
            action = np.random.randint(0, self.n_actions)
        return action

    def learn(self):
        # check to replace target parameters
        if self.learn_step_counter % self.replace_target_iter == 0:
            self.sess.run(self.replace_target_op)
            print('\ntarget_params_replaced\n')

        # sample batch memory from all memory
        if self.memory_counter > self.memory_size:
            sample_index = np.random.choice(self.memory_size, size=self.batch_size)
        else:
            sample_index = np.random.choice(self.memory_counter, size=self.batch_size)
        batch_memory = self.memory[sample_index, :]

        q_next, q_eval = self.sess.run(
            [self.q_next, self.q_eval],
            feed_dict={
                self.s_: batch_memory[:, -self.n_features:],  # fixed params
                self.s: batch_memory[:, :self.n_features],  # newest params
            })

        # change q_target w.r.t q_eval's action
        q_target = q_eval.copy()

        batch_index = np.arange(self.batch_size, dtype=np.int32)
        eval_act_index = batch_memory[:, self.n_features].astype(int)
        reward = batch_memory[:, self.n_features + 1]

        q_target[batch_index, eval_act_index] = reward + self.gamma * np.max(q_next, axis=1)

        # train eval network
        _, self.cost = self.sess.run([self._train_op, self.loss],
                                     feed_dict={self.s: batch_memory[:, :self.n_features],
                                                self.q_target: q_target})
        self.cost_his.append(self.cost)

        # increasing epsilon
        self.epsilon = self.epsilon + self.epsilon_increment if self.epsilon < self.epsilon_max else self.epsilon_max
        self.learn_step_counter += 1

    def plot_cost(self):
        import matplotlib.pyplot as plt
        plt.plot(np.arange(len(self.cost_his)), self.cost_his)
        plt.ylabel('Cost')
        plt.xlabel('training steps')
        plt.show()

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92
  • 93
  • 94
  • 95
  • 96
  • 97
  • 98
  • 99
  • 100
  • 101
  • 102
  • 103
  • 104
  • 105
  • 106
  • 107
  • 108
  • 109
  • 110
  • 111
  • 112
  • 113
  • 114
  • 115
  • 116
  • 117
  • 118
  • 119
  • 120
  • 121
  • 122
  • 123
  • 124
  • 125
  • 126
  • 127
  • 128
  • 129
  • 130
  • 131
  • 132
  • 133
  • 134
  • 135
  • 136
  • 137
  • 138
  • 139
  • 140
  • 141
  • 142
  • 143
  • 144
  • 145
  • 146
  • 147
  • 148
  • 149
  • 150
  • 151
  • 152
  • 153
  • 154
  • 155
  • 156
  • 157
  • 158
  • 159
  • 160
  • 161
  • 162
  • 163
  • 164
  • 165
  • 166
  • 167
  • 168
  • 169
  • 170
  • 171
  • 172
  • 173
  • 174
  • 175
  • 176
  • 177
  • 178
  • 179
  • 180
  • 181
  • 182
  • 183
  • 184
  • 185
  • 186
  • 187
  • 188
  • 189
  • 190
  • 191
  • 192
  • 193
  • 194
  • 195
  • 196
  • 197
  • 198
  • 199
  • 200
  • 201
  • 202
  • 203
  • 204
  • 205
  • 206
  • 207
  • 208
  • 209
  • 210
  • 211
  • 212
  • 213
  • 214
  • 215
  • 216
  • 217
  • 218
  • 219
  • 220
  • 221
  • 222
  • 223
  • 224
声明：本文内容由网友自发贡献，不代表【wpsshop博客】立场，版权归原作者所有，本站不承担相应法律责任。如您发现有侵权的内容，请联系我们。转载请注明出处：https://www.wpsshop.cn/w/笔触狂放9/article/detail/295879
推荐阅读
相关标签

Copyright © 2003-2013 www.wpsshop.cn 版权所有，并保留所有权利。

  

闽ICP备14008679号