使用PPO算法玩“超级马里奥兄弟”

通过本案例的学习和课后作业的练习：

1. 了解PPO算法的基本概念
2. 了解如何基于PPO训练一个小游戏
3. 了解强化学习训练推理游戏的整体流程

你也可以将本案例相关的 ipynb 学习笔记分享到 AI Gallery Notebook 版块获得成长值，分享方法请查看此文档。

案例内容介绍

在此教程中，我们利用PPO算法来玩“Super Mario Bros”（超级马里奥兄弟）。目前来看，对于绝大部分关卡，智能体都可以在1500个episode内学会过关，您可以在超参数栏输入您想要的游戏关卡和训练算法超参数。

整体流程：创建马里奥环境->构建PPO算法->训练->推理->可视化效果

PPO算法的基本结构

PPO算法有两种主要形式：PPO-Penalty和PPO-Clip(PPO2)。在这里，我们讨论PPO-Clip（OpenAI使用的主要形式）。 PPO的主要特点如下：

• PPO属于on-policy算法

• PPO同时适用于离散和连续的动作空间

• 损失函数 PPO-Clip算法最精髓的地方就是加入了一项比例用以描绘新老策略的差异,通过超参数ϵ限制策略的更新步长：
  

• 更新策略：
  

• 探索策略 PPO采用随机探索策略。

• 优势函数 表示在状态s下采取动作a，相较于其他动作有多少优势，如果>0,则当前动作比平均动作好，反之，则差
  

PPO论文

超级马里奥兄弟游戏环境简介

《超级马里奥兄弟》，是任天堂公司开发并于1985年出品的著名横版过关游戏，游戏的目标在于游历蘑菇王国，并从大反派酷霸王的魔掌里救回桃花公主。马力奥可以在游戏世界收集散落各处的金币，或者敲击特殊的砖块，获得其中的金币或特殊道具。这里一共有8大关（world），每大关有4小关（stage）。

注意事项

1. 本案例运行环境为 Pytorch-1.0.0，且需使用 GPU 运行，请查看《ModelAtrs JupyterLab 硬件规格使用指南》了解切换硬件规格的方法；

2. 如果您是第一次使用 JupyterLab，请查看《ModelAtrs JupyterLab使用指导》了解使用方法；

3. 如果您在使用 JupyterLab 过程中碰到报错，请参考《ModelAtrs JupyterLab常见问题解决办法》尝试解决问题。

实验步骤

1. 程序初始化

第1步：安装基础依赖

!pip install -U pip
!pip install gym==0.19.0
!pip install tqdm==4.48.0
!pip install nes-py==8.1.0
!pip install gym-super-mario-bros==7.3.2

import os
import shutil
import subprocess as sp
from collections import deque

import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.multiprocessing as _mp
from torch.distributions import Categorical
import torch.multiprocessing as mp
from nes_py.wrappers import JoypadSpace
import gym_super_mario_bros
from gym.spaces import Box
from gym import Wrapper
from gym_super_mario_bros.actions import SIMPLE_MOVEMENT, COMPLEX_MOVEMENT, RIGHT_ONLY
import cv2
import matplotlib.pyplot as plt
from IPython import display

import moxing as mox

2. 训练参数初始化

该部分参数可以自己调整，以训练出更好的效果

opt={
    "world": 1,                # 可选大关：1,2,3,4,5,6,7,8
    "stage": 1,                # 可选小关：1,2,3,4 
    "action_type": "simple",   # 动作类别："simple"，"right_only", "complex"
    'lr': 1e-4,                # 建议学习率：1e-3，1e-4, 1e-5，7e-5
    'gamma': 0.9,              # 奖励折扣
    'tau': 1.0,                # GAE参数
    'beta': 0.01,              # 熵系数
    'epsilon': 0.2,            # PPO的Clip系数
    'batch_size': 16,          # 经验回放的batch_size
    'max_episode':10,          # 最大训练局数
    'num_epochs': 10,          # 每条经验回放次数
    "num_local_steps": 512,    # 每局的最大步数
    "num_processes": 8,        # 训练进程数，一般等于训练机核心数
    "save_interval": 5,        # 每{}局保存一次模型
    "log_path": "./log",       # 日志保存路径
    "saved_path": "./model",   # 训练模型保存路径
    "pretrain_model": True,    # 是否加载预训练模型，目前只提供1-1关卡的预训练模型，其他需要从零开始训练
    "episode":5
}

3. 创建环境

结束标志：

• 胜利：mario到达本关终点

• 失败：mario受到敌人的伤害、坠入悬崖或者时间用完

奖励函数：

• 得分：收集金币、踩扁敌人、结束时夺旗

• 扣分：受到敌人伤害、掉落悬崖、结束时未夺旗

# 创建环境
def create_train_env(world, stage, actions, output_path=None):
    # 创建基础环境
    env = gym_super_mario_bros.make("SuperMarioBros-{}-{}-v0".format(world, stage))

    env = JoypadSpace(env, actions)
    # 对环境自定义
    env = CustomReward(env, world, stage, monitor=None)
    env = CustomSkipFrame(env)
    return env
# 对原始环境进行修改，以获得更好的训练效果
class CustomReward(Wrapper):
    def __init__(self, env=None, world=None, stage=None, monitor=None):
        super(CustomReward, self).__init__(env)
        self.observation_space = Box(low=0, high=255, shape=(1, 84, 84))
        self.curr_score = 0
        self.current_x = 40
        self.world = world
        self.stage = stage
        if monitor:
            self.monitor = monitor
        else:
            self.monitor = None

    def step(self, action):
        state, reward, done, info = self.env.step(action)
        if self.monitor:
            self.monitor.record(state)
        state = process_frame(state)
        reward += (info["score"] - self.curr_score) / 40.
        self.curr_score = info["score"]
        if done:
            if info["flag_get"]:
                reward += 50
            else:
                reward -= 50
        if self.world == 7 and self.stage == 4:
            if (506 <= info["x_pos"] <= 832 and info["y_pos"] > 127) or (
                    832 < info["x_pos"] <= 1064 and info["y_pos"] < 80) or (
                    1113 < info["x_pos"] <= 1464 and info["y_pos"] < 191) or (
                    1579 < info["x_pos"] <= 1943 and info["y_pos"] < 191) or (
                    1946 < info["x_pos"] <= 1964 and info["y_pos"] >= 191) or (
                    1984 < info["x_pos"] <= 2060 and (info["y_pos"] >= 191 or info["y_pos"] < 127)) or (
                    2114 < info["x_pos"] < 2440 and info["y_pos"] < 191) or info["x_pos"] < self.current_x - 500:
                reward -= 50
                done = True
        if self.world == 4 and self.stage == 4:
            if (info["x_pos"] <= 1500 and info["y_pos"] < 127) or (
                    1588 <= info["x_pos"] < 2380 and info["y_pos"] >= 127):
                reward = -50
                done = True

        self.current_x = info["x_pos"]
        return state, reward / 10., done, info

    def reset(self):
        self.curr_score = 0
        self.current_x = 40
        return process_frame(self.env.reset())
class MultipleEnvironments:
    def __init__(self, world, stage, action_type, num_envs, output_path=None):
        self.agent_conns, self.env_conns = zip(*[mp.Pipe() for _ in range(num_envs)])
        if action_type == "right_only":
            actions = RIGHT_ONLY
        elif action_type == "simple":
            actions = SIMPLE_MOVEMENT
        else:
            actions = COMPLEX_MOVEMENT
        self.envs = [create_train_env(world, stage, actions, output_path=output_path) for _ in range(num_envs)]
        self.num_states = self.envs[0].observation_space.shape[0]
        self.num_actions = len(actions)
        for index in range(num_envs):
            process = mp.Process(target=self.run, args=(index,))
            process.start()
            self.env_conns[index].close()

    def run(self, index):
        self.agent_conns[index].close()
        while True:
            request, action = self.env_conns[index].recv()
            if request == "step":
                self.env_conns[index].send(self.envs[index].step(action.item()))
            elif request == "reset":
                self.env_conns[index].send(self.envs[index].reset())
            else:
                raise NotImplementedError
def process_frame(frame):
    if frame is not None:
        frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
        frame = cv2.resize(frame, (84, 84))[None, :, :] / 255.
        return frame
    else:
        return np.zeros((1, 84, 84))
class CustomSkipFrame(Wrapper):
    def __init__(self, env, skip=4):
        super(CustomSkipFrame, self).__init__(env)
        self.observation_space = Box(low=0, high=255, shape=(skip, 84, 84))
        self.skip = skip
        self.states = np.zeros((skip, 84, 84), dtype=np.float32)

    def step(self, action):
        total_reward = 0
        last_states = []
        for i in range(self.skip):
            state, reward, done, info = self.env.step(action)
            total_reward += reward
            if i >= self.skip / 2:
                last_states.append(state)
            if done:
                self.reset()
                return self.states[None, :, :, :].astype(np.float32), total_reward, done, info
        max_state = np.max(np.concatenate(last_states, 0), 0)
        self.states[:-1] = self.states[1:]
        self.states[-1] = max_state
        return self.states[None, :, :, :].astype(np.float32), total_reward, done, info

    def reset(self):
        state = self.env.reset()
        self.states = np.concatenate([state for _ in range(self.skip)], 0)
        return self.states[None, :, :, :].astype(np.float32)

4. 定义神经网络

神经网络结构包含四层卷积网络和一层全连接网络，提取的特征输入critic层和actor层，分别输出value值和动作概率分布。

class Net(nn.Module):
    def __init__(self, num_inputs, num_actions):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(num_inputs, 32, 3, stride=2, padding=1)
        self.conv2 = nn.Conv2d(32, 32, 3, stride=2, padding=1)
        self.conv3 = nn.Conv2d(32, 32, 3, stride=2, padding=1)
        self.conv4 = nn.Conv2d(32, 32, 3, stride=2, padding=1)
        self.linear = nn.Linear(32 * 6 * 6, 512)
        self.critic_linear = nn.Linear(512, 1)
        self.actor_linear = nn.Linear(512, num_actions)
        self._initialize_weights()

    def _initialize_weights(self):
        for module in self.modules():
            if isinstance(module, nn.Conv2d) or isinstance(module, nn.Linear):
                nn.init.orthogonal_(module.weight, nn.init.calculate_gain('relu'))
                nn.init.constant_(module.bias, 0)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        x = F.relu(self.conv2(x))
        x = F.relu(self.conv3(x))
        x = F.relu(self.conv4(x))
        x = self.linear(x.view(x.size(0), -1))
        return self.actor_linear(x), self.critic_linear(x)

6. 训练模型

训练10 Episode，耗时约5分钟

train(opt)

加载预训练模型
Episode: 1. Total loss: 1.1230244636535645
Episode: 2. Total loss: 2.553663730621338
Episode: 3. Total loss: 1.768389344215393
Episode: 4. Total loss: 1.6962862014770508
Episode: 5. Total loss: 1.0912611484527588
Episode: 6. Total loss: 1.6626232862472534
Episode: 7. Total loss: 1.9952025413513184
Episode: 8. Total loss: 1.2410558462142944
Episode: 9. Total loss: 1.3711413145065308
Episode: 10. Total loss: 1.2155205011367798

7. 使用模型推理游戏

定义推理函数

def infer(opt):
    if torch.cuda.is_available():
        torch.cuda.manual_seed(123)
    else:
        torch.manual_seed(123)
    if opt['action_type'] == "right":
        actions = RIGHT_ONLY
    elif opt['action_type'] == "simple":
        actions = SIMPLE_MOVEMENT
    else:
        actions = COMPLEX_MOVEMENT
    env = create_train_env(opt['world'], opt['stage'], actions)
    model = Net(env.observation_space.shape[0], len(actions))
    if torch.cuda.is_available():
        model.load_state_dict(torch.load("{}/ppo_super_mario_bros_{}_{}_{}".format(opt['saved_path'],opt['world'], opt['stage'],opt['episode'])))
        model.cuda()
    else:
        model.load_state_dict(torch.load("{}/ppo_super_mario_bros_{}_{}_{}".format(opt['saved_path'], opt['world'], opt['stage'],opt['episode']),
                                         map_location=torch.device('cpu')))
    model.eval()
    state = torch.from_numpy(env.reset())
    
    plt.figure(figsize=(10,10))
    img = plt.imshow(env.render(mode='rgb_array'))
    
    while True:
        if torch.cuda.is_available():
            state = state.cuda()
        logits, value = model(state)
        policy = F.softmax(logits, dim=1)
        action = torch.argmax(policy).item()
        state, reward, done, info = env.step(action)
        state = torch.from_numpy(state)
        
        img.set_data(env.render(mode='rgb_array')) # just update the data
        display.display(plt.gcf())
        display.clear_output(wait=True)
        
        if info["flag_get"]:
            print("World {} stage {} completed".format(opt['world'], opt['stage']))
            break
            
        if done and info["flag_get"] is False:
            print('Game Failed')
            break

infer(opt)

8. 作业¶

1. 请你调整步骤2中的训练参数，重新训练一个模型，使它在游戏中获得更好的表现。