# Reinforcement_learning

**Repository Path**: jiang_maowei/reinforcement_learning

## Basic Information

- **Project Name**: Reinforcement_learning
- **Description**: 自己跑过的强化学习一些简单代码和入门教程
- **Primary Language**: Python
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2021-12-16
- **Last Updated**: 2022-07-01

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Reinforcement_learning

自己跑过的强化学习一些简单代码和入门教程，主要来自教程 https://mofanpy.com/tutorials/machine-learning/reinforcement-learning/ 以及 https://github.com/MorvanZhou/Reinforcement-learning-with-tensorflow

## 1_command_line_Q_learning
利用Q Learning完成一维空间的寻宝任务，基于Off-policy

## 2_Q_Learning_maze
4 x 4 方格，跟1_command_line_Q_learning很像只不过是二维情况，而且会有负反馈

## 3_DQN_CartPole
DQN实现倒立摆，利用OpenAI Gym环境
```
pip install gym
pip install pyglet
```