AlphaZero is a game-playing algorithm that uses artificial intelligence and machine learning techniques to learn how to play board games at a superhuman level.

We just published a machine learning course on the channel that will teach you how to build AlphaZero from scratch.

Robert Förster created this course. He is a student from Germany who is focused on machine learning.

The video course teaches how to code an AlphaZero algorithm from scratch to play Tic Tac Toe and Connect Four. The course is divided into ten sections, starting with an introduction to the course and an overview of the AlphaZero algorithm.

In Part 1 of the overview, the course explains the Monte Carlo Tree Search (MCTS) algorithm used in AlphaZero, and in Part 2, it covers the AlphaMCTS algorithm, which is a modification of MCTS used in AlphaZero.

The course then moves on to Chapter 1, where the code for Tic Tac Toe is implemented using AlphaZero. In Chapter 2, the course explains how to implement MCTS for Tic Tac Toe. Chapter 3 covers the model used in AlphaZero, which is a neural network that learns to play the game.

In Chapter 4, the course covers AlphaMCTS, which combines AlphaZero's neural network with MCTS. Chapter 5 covers AlphaSelfPlay, which is the process of having AlphaZero play against itself to improve its gameplay.

Chapter 6 covers AlphaTrain, which is the process of training the neural network using the data generated from AlphaSelfPlay. In Chapter 7, the course covers AlphaTweaks, which are modifications made to the algorithm to improve performance.

Chapter 8 covers Connect Four, where the course implements AlphaZero to play the game. Chapter 9 covers AlphaParallel, which is the process of training multiple neural networks simultaneously. Finally, Chapter 10 covers Eval, which is the process of evaluating the performance of the trained neural network.

Overall, the video course provides a comprehensive overview of how to code an AlphaZero algorithm from scratch to play two different games, Tic Tac Toe and Connect Four.

Watch the full course on the YouTube channel (4-hour watch).