Introduction

Application

• Game

  • Backgammon game
    • 10 - 20 game state
    • TD-Gammon: the 1st NN that advanced the theroy of Backgammon by discovering strategies previously unkown
  • Atari breakout
  • Dota2
    • OpenAI
  • AlphaGo

• Robotics

  • Walk
    • Producing flexible behaviours in simulated environments
  • Driving

• Other:

  • Biology
  • Telecommunications
    • Low Power Wireless Communication via Reinforcement Learning
  • Finance
    • Introduction to Learning to Trade with Reinforcement Learning
  • Inventory management
    • Inventory management in supply chains: a reinforcement learning approach

Exploration-Exploitation Dilemma

Managing the Exploration-Exploitation Dilemma, image from 52Aces

• ← Previous
Next →