/Reinforcement-Learning-in-Robotics

This is a private learning repository for reinforcement learning techniques used in robotics.

Primary LanguageHTMLMIT LicenseMIT

Reinforcement-Learning-in-Robotics

This is a private learning repository for reinforcement learning techniques used in robotics.

Related Works

Reinforcement Leaning Foundation

  1. 神经网络基础: 反向传播推导与卷积公式 [Zhihu] [Github]
  2. 强化学习基础 Ⅰ：马尔可夫与值函数 [Zhihu] [Github]
  3. 强化学习基础 Ⅱ：动态规划，蒙特卡洛，时序差分 [Zhihu] [Github]
  4. 强化学习基础 Ⅲ：on-policy, off-policy & Model-based, Model-free & Rollout [Zhihu] [Github]
  5. 强化学习基础 Ⅳ：State-of-the-art 强化学习经典算法汇总 [Zhihu] [Github]
  6. 强化学习基础 Ⅴ：Q learning 原理与实战 [Zhihu]
  7. 强化学习基础 Ⅵ：DQN 原理与实战 [Zhihu]
  8. 强化学习基础 Ⅶ：Double DQN & Dueling DQN 原理与实战 [Zhihu]
  9. 强化学习基础 Ⅷ：Vanilla Policy Gradient 策略梯度原理与 [Zhihu]
  10. 强化学习基础 Ⅸ：一文读懂 TRPO 原理与实现 [Zhihu]
  11. 强化学习基础 Ⅹ：一文读懂两种 PPO 原理与实现 [zhihu]

Model-based RL

  1. Model-Based RL Ⅰ：Dyna, MVE & STEVE [Zhihu]
  2. Model-Based RL Ⅱ：MBPO原理解读 [Zhihu]
  3. Model-Based RL Ⅲ：从源码读懂PILCO [Zhihu]

Probability in Robotics

  1. 机器人学的概率方法——最大似然估计MLE与最大后验概率估计MAP [Zhihu]
  2. PR Ⅱ：贝叶斯估计/推断及其与MAP的区别 [Zhihu]
  3. PR Ⅲ：从高斯分布到高斯过程、高斯过程回归、贝叶斯优化 [Zhihu]
  4. PR Ⅳ：贝叶斯神经网络 Bayesian Neural Network [Zhihu]
  5. PR Ⅴ：熵、KL散度、交叉熵、JS散度及python实现 [Zhihu]
  6. PR Ⅵ：多元连续高斯分布的KL散度及python实现 [Zhihu]

Meta-Learning

  1. Meta-Learning: An Introduction Ⅰ [Zhihu] [Github]
  2. Meta-Learning: An Introduction Ⅱ [Zhihu] [Github]
  3. Meta-Learning: An Introduction Ⅲ [Zhihu] [Github]

Imitation Learning

  1. 模仿学习(Imitation Learning)入门指南 [Zhihu]
  2. Imitation Learning Ⅱ：DAgger透彻理论分析 [Zhihu]
  3. Imitation Learning Ⅲ：EnsembleDAgger 一种贝叶斯DAgger [Zhihu]

RL from Demonstrations

  1. RLfD Ⅰ：Deep Q-learning from Demonstrations 解读 [Zhihu]
  2. RLfD Ⅱ：Reinforcement Learning from Imperfect Demonstrations under Soft Expert Guidance [Zhihu]

Paper Reading

  1. End-to-End Robotic Reinforcement Learning without Reward Engineering: [Medium] [Github] [Zhihu]
  2. Overcoming Exploration in RL with Demonstrations: [Medium] [Github] [Zhihu]
  3. The Predictron: End-To-End Learning and Planning: [Zhihu] [Github]
  4. IROS2019 Paper速读(一): [Zhihu] [Github]
  5. IROS2019 Paper速读(二): [Zhihu] [Github]
  6. IROS2019 Paper速读(三): [Zhihu] [Github]
  7. IROS2019 Paper速读(四): [Zhihu] [Github]
  8. 【重磅综述】如何在少量尝试下学习机器人强化学习控制 [Zhihu]

Simulator

MuJoCo自定义机器人建模指南 [Zhihu]