2021 HAI Summer RL

한양대학교 인공지능 동아리 HAI에서 2021년 여름에 진행된 강화학습 기초 강의 자료입니다. 예제 자료와 과제는 Numpy와 PyTorch를 사용합니다.

Text Books

• Deep Reinforcement Learning Hands-On - Second Edition (Packt, 2020)
• Reinforcement Learning, An Introduction - Second Edition (MIT Presse, 2018)
  • Korean: 단단한 강화학습 (제이펍, 2020)

Contents

• Week1 - Introduction to RL and OpenAI Gym (7/23) [Lecture] [Colab]

  • Introduction to RL
  • OpenAI Gym

• Week2 - MDP and Cross Entropy Method (7/30) [Lecture] [Code]

  • Markov Decision Process
  • Cross Entropy Method

• Week3 - Bellman Equation and Tabular Learning (8/6) [Lecture] [Code]

  • Bellman Equation
  • Dynamic Programming
  • Tabular Learning

• Week4 - Deep Q-Network (8/13) [Lecture] [PyTorch Practice] [Code]

  • Approximate Solution Methods
  • Deep Q-Network

• Week5 - Policy Gradient (8/20) [Lecture] [Code]

  • Policy Gradient
  • REINFORCE

• Week6 - Actor Critic (8/27) [Lecture] [Code]

  • REINFORCE Issues
  • Actor Critic