Astral Codex Ten

AlphaZero

4 min read

AlphaZero

Article

AlphaZero is a recurring concept in the Astral Codex Ten archive, appearing 2 times across 2 issues between August 06, 2021 and October 03, 2022. The archive places it in contexts such as “DeepMind got their Go AI AlphaZero to try learning chess”; “AlphaZero etc ARE old-fashioned search-tree players”. It most often appears alongside Eliezer Yudkowsky, AGI, AI.

Metadata

  • Category: Concepts
  • Mention count: 2
  • Issue count: 2
  • First seen: August 06, 2021
  • Last seen: October 03, 2022

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

Highlights From The Comments On Acemoglu And AI
August 06, 2021 · Original source
But can the learning algorithm learn to play chess? Yes, extremely well. DeepMind got their Go AI AlphaZero to try learning chess, and it became world champion within a day. Then they asked it to learn a different game called shogi, and it became world champion of that one too. Could AlphaZero learn how to invent new rockets? No, because that’s not the class of problems it knows how to learn about (it’s not a board game where it can play against itself a bunch of times and observe its mistakes). So is the learning algorithm a narrow AI or a general AI? It’s not infinitely narrow - it can learn any board game you throw at it - but it’s not infinitely general either. Certainly it’s more general, smarter, and at least slightly scarier than a polynomial that predicts parole decisions.
Right now a lot of research is going into making things that are slightly more general than AlphaZero. For example, could you get something which, in addition to being able to play any board game, can also play any video game? This turns out to be a really different problem; my understanding is that they’re pretty close but not quite there. What about just games in general? Last week, DeepMind published a paper, Open-Ended Learning Leads To Generally Capable Agents. They created a simulated 3D physical environment, stuck an AI in a simulated body in that environment, and made it go through various obstacle courses and stuff. They found that the knowledge generalized, so that the AI was eventually able to learn to play games they hadn’t taught it, like hide-and-seek and capture-the-flag, coming up with decent strategies on their first attempt based on the general principles it had learned from other things. Where does this place it on the “it’s just an algorithm” vs. “real intelligence” dichotomy?
Inline links: Open-Ended Learning Leads To Generally Capable Agents
CHAI, Assistance Games, And Fully-Updated Deference
October 03, 2022 · Original source
Not sure I follow this [part] at all. Wouldn't the same argument apply to the method described above for "the only way we know how to train modern AIs"? Is Eliezer saying that good old-fashioned rule-based systems never existed and could not exist? Or that perception isn't perfect? . . . AlphaZero etc ARE old-fashioned search-tree players