Reinforcement Learning policy evaluation实现以及OpenAI Gym介绍

最新推荐文章于 2025-07-20 17:47:42 发布

Snail_Walker

最新推荐文章于 2025-07-20 17:47:42 发布

阅读量1.6k

点赞数 2

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/c602273091/article/details/78999056

本文介绍了如何使用Python实现RL Policy Evaluation，并详细解析了OpenAI Gym，包括其作为强化学习工具包的角色，环境抽象，Observation和Spaces的概念。OpenAI Gym提供多种经典控制问题、Atari游戏及棋盘游戏等环境。文章总结了强化学习中的关键概念，并展示了与环境交互的基本步骤。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

RL Policy Evaluation Python实现
OpenAI Gym

RL Policy Evaluation Python实现

根据UCL课程Lecture 3的4x4 grid求policy evaluation的方法，可以写成如下实现：

states = [i for i in range(16)]
values = [0 for _ in range(16)]
actions = ["n", "e", "s", "w"]
ds_actions = {
  
  "n":-1, "e":1, "s":4, "w": -1}
gamma = 1.00

def nextState(s,a):
    next_state = s
    if (s%4 == 0 and a == "w") or (s<4 and a == "n") or  ((s+1)%4 == 0 and a == "e") or (s > 11 and a == "s"):
        pass
    else:
        ds = ds_actions[a]
        next_state = s + ds
    return next_state


def rewardOf(s):
    return 0 if s in [0,15] else -1

def isTerminateState(s):
    return s in [