When it comes to progress, Artificial Intelligence is often under the spotlights because it evolves quickly, and no field is left behind. Even the playing field is concerned with the conception of new computer programs which can compete or even beat professional players, like for example AlphaGo.
AlphaGo is a computer program developed by Deepmind Technologies that plays the board game Go. In 2015, it became the first program to beat a professional Go player, which marked a significant milestone in the evolution of AI.
In 2016, by combining machine learning and graph path search based techniques, it achieved to beat one of the best Go players in the world (although this player was retired from competition).
It was then defeated by AlphaGo Zero, a stronger version which was completely self-taught without learning from human games and scored 100-0 against AlphaGo.
The program then became Alpha Zero in 2017 and is currently the world’s best players among programs and humans in Go, chess and shōgi. In 2019, AlphaZero has been turned into a program known as MuZero which learns without being taught the rules.
3 Responses
Where can I find more information on MuZero?
Hi Rick !
MuZero is pretty cool right?
Here is a good youtube video:
https://www.youtube.com/watch?v=szbvm8aNDxw
if you’re lucky enough to be part of a research institution, you can go deeper :
https://www.nature.com/articles/MuZero
Enjoy!
Very interesting article! I heard also that even the champion Gasparov was beaten by AI during a chess competition! Human will be soon obsolete haha 🙂