Introduction to CFR and implementation of the Tic-Tac-Toe game

Photo by Brett Jordan on Unsplash

We should regret our mistakes and learn from them, but never carry them forward into the future with us.”

Lucy Maud Montgomery

Sometimes regretting is a good way to improve

Photo by Free To Use Sounds on Unsplash



A concrete implementation of Neural Fictitious Self-Play in Leduc Hold’em Poker Game

Photo by Keenan Constance on Unsplash

A gentle approach for developers to decipher the math formula of Reinforcement Learning

Photo by Tine Ivanič on Unsplash

A pragmatic approach to understand policies in Reinforcement Learning

Photo by Philip Wyers on Unsplash


Deep Reinforcement Learning in Imperfect-Information Games

Photo by Paweł Czerwiński on Unsplash


Important Reminders

  • Fictitious Play: is an iterative method that finds Nash Equilibrium in Two Player Zero Sum game. Its problem is that it applies to Normal Form Games which are tabular and do not capture time or sequence. Details can be found in the article “Introduction to Fictitious Play”.
  • Fictitious Self Play: is a method that fixes the problem of the Fictitious Play by integrating time/sequence by using Extensive Form Game. It also uses…

A step forward towards Self Play in Reinforcement Learning

Photo by Markus Spiske on Unsplash


Extensive Form Game

Photo by Mpho Mojapelo on Unsplash

Zero Sum Game

Stitch together the different parts and branches of Reinforcement Learning

Photo by Josh Power on Unsplash

Intuitive approach to understand one of the most important methods in Reinforcement Learning.

Photo by Fatih Kılıç on Unsplash


Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store