> can learn the game by itself without knowing the rules

In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6.

“What are you doing?”, asked Minsky.

“I am training a randomly wired neural net to play Tic-Tac-Toe” Sussman replied.

“Why is the net wired randomly?”, asked Minsky.

“I do not want it to have any preconceptions of how to play”, Sussman said.

Minsky then shut his eyes.

“Why do you close your eyes?”, Sussman asked his teacher.

“So that the room will be empty.”

At that moment, Sussman was enlightened.

http://www.catb.org/jargon/html/koans.html

This "Zero-knowledge" style program, be it AlphaZero or MuZero and whatever, is a bad meme.

Adding in little bits of knowledge, such as liberties counting, ladder-reading, and other tricks, allows KataGo to absolutely crush LeelaZero in training speed and practical play in the game of Go. Its no comparison.

These days, no one studying the game of Go is seriously using LeelaZero anymore. Its an obsolete methodology compared to KataGo.

A more familiar example might be fuzz testing. You could fuzz a program by just passing the fuzzer totally random data to mutate, and it's true that this might find a crash that wouldn't be found any other way. However, most of the time you will have much better results, and achieve them much more quickly, if you pass the fuzzer something that looks like an actual expected input to the program.

Zero-knowledge ML is still interesting if your goal is to indisputably demonstrate the act of learning, just not if your goal is to make the smartest program in the least time.