Deep Q-Network (DQN)理解

本文主要是介绍Deep Q-Network (DQN)理解，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

DQN（Deep Q-Network）是深度强化学习（Deep Reinforcement Learning）的开山之作，将深度学习引入强化学习中，构建了 Perception 到 Decision 的 End-to-end 架构。DQN 最开始由 DeepMind 发表在 NIPS 2013，后来将改进的版本发表在 Nature 2015。

NIPS 2013: Playing Atari with Deep Reinforcement Learning
Nature 2015: Human-level control through deep reinforcement learning

DQN 面临着几个挑战：

深度学习需要大量带标签的训练数据；
强化学习从 scalar reward 进行学习，但是 reward 经常是 sparse, noisy, delayed；
深度学习假设样本数据是独立同分布的，但是强化学习中采样的数据是强相关的

因此，DQN 采用经验回放（Experience Replay）机制，将训练过的数据进行储存到 Replay Buffer 中，以便后续从中随机采样进行训练，好处就是：1. 数据利用率高；2. 减少连续样本的相关性，从而减小方差（variance）。

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

    # total learning stepself.learn_step_counter = 0# initialize zero memory [s, a, r, s_]self.memory = np.zeros((self.memory_size, n_features * 2 + 2))# consist of [target_net, evaluate_net]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:# $ tensorboard --logdir=logstf.summary.FileWriter("logs/", self.sess.graph)self.sess.run(tf.global_variables_initializer())self.cost_his = []def _build_net(self):# ------------------ build evaluate_net ------------------self.s = tf.placeholder(tf.float32, [None, self.n_features], name='s')  # inputself.q_target = tf.placeholder(tf.float32, [None, self.n_actions], name='Q_target')  # for calculating losswith tf.variable_scope('eval_net'):# c_names(collections_names) are the collections to store variablesc_names = ['eval_net_params', tf.GraphKeys.GLOBAL_VARIABLES] n_l1 = 10w_initializer = tf.random_normal_initializer(0., 0.3)b_initializer = tf.constant_initializer(0.1)# first layer. collections is used later when assign to target netwith 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)# second layer. collections is used later when assign to target netwith 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) + b2with 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)# ------------------ build target_net ------------------self.s_ = tf.placeholder(tf.float32, [None, self.n_features], name='s_')    # inputwith tf.variable_scope('target_net'):# c_names(collections_names) are the collections to store variablesc_names = ['target_net_params', tf.GraphKeys.GLOBAL_VARIABLES]# first layer. collections is used later when assign to target netwith 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)# second layer. collections is used later when assign to target netwith 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) + b2def store_transition(self, s, a, r, s_):if not hasattr(self, 'memory_counter'):self.memory_counter = 0transition = np.hstack((s, [a, r], s_))# replace the old memory with new memoryindex = self.memory_counter % self.memory_sizeself.memory[index, :] = transitionself.memory_counter += 1def choose_action(self, observation):# to have batch dimension when feed into tf placeholderobservation = observation[np.newaxis, :]if np.random.uniform() < self.epsilon:# forward feed the observation and get q value for every actionsactions_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 actiondef learn(self):# check to replace target parametersif 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 memoryif 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 paramsself.s: batch_memory[:, :self.n_features],  # newest params})# change q_target w.r.t q_eval's actionq_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)"""For example in this batch I have 2 samples and 3 actions:q_eval =[[1, 2, 3],[4, 5, 6]]q_target = q_eval =[[1, 2, 3],[4, 5, 6]]Then change q_target with the real q_target value w.r.t the q_eval's action.For example in:sample 0, I took action 0, and the max q_target value is -1;sample 1, I took action 2, and the max q_target value is -2:q_target =[[-1, 2, 3],[4, 5, -2]]So the (q_target - q_eval) becomes:[[(-1)-(1), 0, 0],[0, 0, (-2)-(6)]]We then backpropagate this error w.r.t the corresponding action to network,leave other action as error=0 cause we didn't choose it."""# 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 epsilonself.epsilon = self.epsilon + self.epsilon_increment if self.epsilon < self.epsilon_max else self.epsilon_maxself.learn_step_counter += 1

在DQN中增强学习Q-Learning算法和深度学习的SGD训练是同步进行的！

通过Q-Learning获取无限量的训练样本，然后对神经网络进行训练。

样本的获取关键是计算y，也就是标签。

这篇关于Deep Q-Network (DQN)理解的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！

---