AlphaZero learns human concepts

Deepmind's AlphaZero is considered an AI milestone. A new paper examines how exactly the AI system has learned chess - and how close it is to humans.

Reimagining Chess with AlphaZero, February 2022

Assessing Game Balance with AlphaZero: Exploring Alternative Rule

Lessons from AlphaZero for Optimal, by Dimitri P. Bertsekas

DeepMind AlphaTensor: The delicate balance between human and

AlphaZero learns to use the positive concepts; Model Code = (left

Simple Alpha Zero

Superhuman AI Triumphs Playing the Toughest Board Games

Acquisition of Chess Knowledge in AlphaZero

Top dog: AlphaZero / AlphaStar. AlphaZero and AlphaStar are two

AlphaZero AI Shows Signs of Developing a Sense of Intuition - NullTX

On its own, in just a few hours of experimental self-play

AlphaZero, a novel Reinforcement Learning Algorithm, in JavaScript

The human side of AI for chess - Microsoft Research