RLearning: Short guides to reinforcement learning

Short guides with tricks and tips to understand and apply reinforcement learning.

• 0-1 Introduction
  
  
• 1-1 Multi-Armed Bandits
  
  
• 1-2 epsilon-greedy
  
  
• 1-3 upper confidence bound
  
  
• 1-4 Thompson sampling
  
  
• 1-5 Inference with batched bandits
  
  
• 2-1 Markov Processes
  
  
• 2-2 Markov Decision Processes
  
  
• 2-3 Intro to Value Iteration with Bellman Equation
  
  
• 2-4 Value Iteration Technicalities
  
  
• 2-5 Policy Iteration
  
  
• 3-1 Overview of Reinforcement Learning
  
  
• 3-2 Monte Carlo Learning
  
  
• 3-3 Temporal Difference Learning
  
  
• 3-4 Q-Learning
  
  
• 4-1 Neural Networks
  
  
• 4-2 Deep Neural Networks
  
  
• 4-3 Deep Q-Learning
  
  
• 4-4 Policy Gradient Methods
  
  
• 4-5 Actor Critic Algorithms
  
  

Python programs and assignments with tricks and tipps to understand and apply reinforcement learning.

• Assignment 1: Multi-Armed Bandit Portfolio
  
  
• Assignment 2: Value Iteration and Q-Learning
  
  

Watch video recordings discussing the slides on youtube.

All LaTeX files and codes available on the RLearning github page.

Contact

Davud Rostam-Afschar
Universität Mannheim
rostam-afschar@uni-mannheim.de