从零使用强化学习训练AI玩儿游戏(4)——使用Sarsa

            本文目前主要是写给自己的一个笔记，接下来这段时间会逐步记录我是怎么通过学习使用TensorFlow+Keras训练神经网络自己玩儿游戏，如果能间接帮助到他人就最好不过了，不喜勿喷。

        上一篇我们已经使用了Q-learning来玩儿一个简单Gym游戏，由于游戏的状态不一定，导致一直更新Q表，所以这个算法在玩儿这种游戏的时候毫无用处，可以玩走迷宫这种。

        这一篇我们就来使用Sarsa玩儿游戏，Sarsa是Q-learning的一个进化版本，莫烦大神的视频详细介绍了Sarsa

        Q-learning是说到不一定做到的类型所以是off-policy，Sarsa是说道一定做到类型所以是on-policy，如下图所示是两个算法直接的区别。

        在图片中可以看到Q-learning的take action直接走和choose action从Q表中选择中间加的是更新Q表，也就是说下一步怎么走，要等更新完Q表后才能做决定，其中更新Q表只是假设这么走了会是什么情况。

        Sarsa的的take action直接走和choose action从Q表中选择中间没有加东西，就是走了这步先不更新Q表直接走下一步，然后再更新Q表。

        Sarsa其实和Q-learning一样，也是针对更新Q表的算法，所以也是不适用于昨天gym上的游戏的，今天我们就用莫烦大神的宝藏游戏来看一下。run_this.py

from maze_env import Maze
from RL_brain import SarsaTable


def update():
    for episode in range(100):
        # initial observation
        observation = env.reset()

        # RL choose action based on observation
        action = RL.choose_action(str(observation))

        while True:
            # fresh env
            env.render()

            # RL take action and get next observation and reward
            observation_, reward, done = env.step(action)

            # RL choose action based on next observation
            action_ = RL.choose_action(str(observation_))

            # RL learn from this transition (s, a, r, s, a) ==> Sarsa
            RL.learn(str(observation), action, reward, str(observation_), action_)

            # swap observation and action
            observation = observation_
            action = action_

            # break while loop when end of this episode
            if done:
                break

    # end of game
    print('game over')
    env.destroy()

if __name__ == "__main__":
    env = Maze()
    RL = SarsaTable(actions=list(range(env.n_actions)))

    env.after(100, update)
    env.mainloop()

import numpy as np
import pandas as pd


class RL(object):
    def __init__(self, action_space, learning_rate=0.01, reward_decay=0.9, e_greedy=0.9):
        self.actions = action_space  # a list
        self.lr = learning_rate
        self.gamma = reward_decay
        self.epsilon = e_greedy

        self.q_table = pd.DataFrame(columns=self.actions, dtype=np.float64)

    def check_state_exist(self, state):
        if state not in self.q_table.index:
            # append new state to q table
            self.q_table = self.q_table.append(
                pd.Series(
                    [0]*len(self.actions),
                    index=self.q_table.columns,
                    name=state,
                )
            )

    def choose_action(self, observation):
        self.check_state_exist(observation)
        # action selection
        if np.random.rand() < self.epsilon:
            # choose best action
            state_action = self.q_table.loc[observation, :]
            state_action = state_action.reindex(np.random.permutation(state_action.index))     # some actions have same value
            action = state_action.idxmax()
        else:
            # choose random action
            action = np.random.choice(self.actions)
        return action

    def learn(self, *args):
        pass


# off-policy
class QLearningTable(RL):
    def __init__(self, actions, learning_rate=0.01, reward_decay=0.9, e_greedy=0.9):
        super(QLearningTable, self).__init__(actions, learning_rate, reward_decay, e_greedy)

    def learn(self, s, a, r, s_):
        self.check_state_exist(s_)
        q_predict = self.q_table.loc[s, a]
        if s_ != 'terminal':
            q_target = r + self.gamma * self.q_table.loc[s_, :].max()  # next state is not terminal
        else:
            q_target = r  # next state is terminal
        self.q_table.loc[s, a] += self.lr * (q_target - q_predict)  # update


# on-policy
class SarsaTable(RL):

    def __init__(self, actions, learning_rate=0.01, reward_decay=0.9, e_greedy=0.9):
        super(SarsaTable, self).__init__(actions, learning_rate, reward_decay, e_greedy)

    def learn(self, s, a, r, s_, a_):
        self.check_state_exist(s_)
        q_predict = self.q_table.loc[s, a]
        if s_ != 'terminal':
            q_target = r + self.gamma * self.q_table.loc[s_, a_]  # next state is not terminal
        else:
            q_target = r  # next state is terminal
        self.q_table.loc[s, a] += self.lr * (q_target - q_predict)  # update

import numpy as np
import time
import sys
if sys.version_info.major == 2:
    import Tkinter as tk
else:
    import tkinter as tk


UNIT = 40   # pixels
MAZE_H = 4  # grid height
MAZE_W = 4  # grid width


class Maze(tk.Tk, object):
    def __init__(self):
        super(Maze, self).__init__()
        self.action_space = ['u', 'd', 'l', 'r']
        self.n_actions = len(self.action_space)
        self.title('maze')
        self.geometry('{0}x{1}'.format(MAZE_H * UNIT, MAZE_H * UNIT))
        self._build_maze()

    def _build_maze(self):
        self.canvas = tk.Canvas(self, bg='white',
                           height=MAZE_H * UNIT,
                           width=MAZE_W * UNIT)

        # create grids
        for c in range(0, MAZE_W * UNIT, UNIT):
            x0, y0, x1, y1 = c, 0, c, MAZE_H * UNIT
            self.canvas.create_line(x0, y0, x1, y1)
        for r in range(0, MAZE_H * UNIT, UNIT):
            x0, y0, x1, y1 = 0, r, MAZE_H * UNIT, r
            self.canvas.create_line(x0, y0, x1, y1)

        # create origin
        origin = np.array([20, 20])

        # hell
        hell1_center = origin + np.array([UNIT * 2, UNIT])
        self.hell1 = self.canvas.create_rectangle(
            hell1_center[0] - 15, hell1_center[1] - 15,
            hell1_center[0] + 15, hell1_center[1] + 15,
            fill='black')
        # hell
        hell2_center = origin + np.array([UNIT, UNIT * 2])
        self.hell2 = self.canvas.create_rectangle(
            hell2_center[0] - 15, hell2_center[1] - 15,
            hell2_center[0] + 15, hell2_center[1] + 15,
            fill='black')

        # create oval
        oval_center = origin + UNIT * 2
        self.oval = self.canvas.create_oval(
            oval_center[0] - 15, oval_center[1] - 15,
            oval_center[0] + 15, oval_center[1] + 15,
            fill='yellow')

        # create red rect
        self.rect = self.canvas.create_rectangle(
            origin[0] - 15, origin[1] - 15,
            origin[0] + 15, origin[1] + 15,
            fill='red')

        # pack all
        self.canvas.pack()

    def reset(self):
        self.update()
        time.sleep(0.5)
        self.canvas.delete(self.rect)
        origin = np.array([20, 20])
        self.rect = self.canvas.create_rectangle(
            origin[0] - 15, origin[1] - 15,
            origin[0] + 15, origin[1] + 15,
            fill='red')
        # return observation
        return self.canvas.coords(self.rect)

    def step(self, action):
        s = self.canvas.coords(self.rect)
        base_action = np.array([0, 0])
        if action == 0:   # up
            if s[1] > UNIT:
                base_action[1] -= UNIT
        elif action == 1:   # down
            if s[1] < (MAZE_H - 1) * UNIT:
                base_action[1] += UNIT
        elif action == 2:   # right
            if s[0] < (MAZE_W - 1) * UNIT:
                base_action[0] += UNIT
        elif action == 3:   # left
            if s[0] > UNIT:
                base_action[0] -= UNIT

        self.canvas.move(self.rect, base_action[0], base_action[1])  # move agent

        s_ = self.canvas.coords(self.rect)  # next state

        # reward function
        if s_ == self.canvas.coords(self.oval):
            reward = 1
            done = True
            s_ = 'terminal'
        elif s_ in [self.canvas.coords(self.hell1), self.canvas.coords(self.hell2)]:
            reward = -1
            done = True
            s_ = 'terminal'
        else:
            reward = 0
            done = False

        return s_, reward, done

    def render(self):
        time.sleep(0.1)
        self.update()

        运行的效果如上图所示，可以明显的看出Sarsa是一个很保守的算法，很怕掉进坑里面，而Q-learning是一个很勇敢的算法，能很快的找到最短路径。

        接下来我们看一下Sarsa(λ)，Sarsa-lambda算法，

        Sarsa-lambda简单的说就是在Sarsa上增加了一个跟Q表一样大小的表，这个表我们先简称问E表，他会在每一次访问到这个状态值得时候给这个状态值+1或者=1，然后每一步这个E表的所有值都会有一个λ大小的衰减，换句话说这个表就是一个会记录你走过的地方的表，并且你走的越远你以前的那些步的数就越小。

        通过这样的方法Sarsa-lambda就可以实现每次找到宝藏的时候对所有走过的路径都实现Q表的更新。

        接下来看一下代码，只需要在Sarsa算法上修改一下RL-brain这个文件

import numpy as np
import pandas as pd


class RL(object):
    def __init__(self, action_space, learning_rate=0.01, reward_decay=0.9, e_greedy=0.9):
        self.actions = action_space  # a list
        self.lr = learning_rate
        self.gamma = reward_decay
        self.epsilon = e_greedy

        self.q_table = pd.DataFrame(columns=self.actions, dtype=np.float64)

    def check_state_exist(self, state):
        if state not in self.q_table.index:
            # append new state to q table
            self.q_table = self.q_table.append(
                pd.Series(
                    [0]*len(self.actions),
                    index=self.q_table.columns,
                    name=state,
                )
            )

    def choose_action(self, observation):
        self.check_state_exist(observation)
        # action selection
        if np.random.rand() < self.epsilon:
            # choose best action
            state_action = self.q_table.loc[observation, :]
            state_action = state_action.reindex(np.random.permutation(state_action.index))     # some actions have same value
            action = state_action.idxmax()
        else:
            # choose random action
            action = np.random.choice(self.actions)
        return action

    def learn(self, *args):
        pass


# backward eligibility traces
class SarsaLambdaTable(RL):
    def __init__(self, actions, learning_rate=0.01, reward_decay=0.9, e_greedy=0.9, trace_decay=0.9):
        super(SarsaLambdaTable, self).__init__(actions, learning_rate, reward_decay, e_greedy)

        # backward view, eligibility trace.
        self.lambda_ = trace_decay
        self.eligibility_trace = self.q_table.copy()

    def check_state_exist(self, state):
        if state not in self.q_table.index:
            # append new state to q table
            to_be_append = pd.Series(
                    [0] * len(self.actions),
                    index=self.q_table.columns,
                    name=state,
                )
            self.q_table = self.q_table.append(to_be_append)

            # also update eligibility trace
            self.eligibility_trace = self.eligibility_trace.append(to_be_append)

    def learn(self, s, a, r, s_, a_):
        self.check_state_exist(s_)
        q_predict = self.q_table.loc[s, a]
        if s_ != 'terminal':
            q_target = r + self.gamma * self.q_table.loc[s_, a_]  # next state is not terminal
        else:
            q_target = r  # next state is terminal
        error = q_target - q_predict

        # increase trace amount for visited state-action pair

        # Method 1:
        # self.eligibility_trace.loc[s, a] += 1

        # Method 2:
        self.eligibility_trace.loc[s, :] *= 0
        self.eligibility_trace.loc[s, a] = 1

        # Q update
        self.q_table += self.lr * error * self.eligibility_trace

        # decay eligibility trace after update
        self.eligibility_trace *= self.gamma*self.lambda_

        还是之前迷宫的环境，可以明显的看出训练的时间减少了，但是每次的路径会变得很长，只要第一次找到宝藏，那很有可能接下来都是这个路线了。

        下一篇我们就会学习我最喜欢的DQN，也就是这几年最流行的强化学习算法了，然后我们可以搭建自己喜欢的神经网络来玩Gym上的游戏，也终于能达到我们标题的要求了，哈哈哈。