4 ms·

> several orders of magnitude more than a human master encounters Not if you account for millions of years of evolution that got your nervous system to its cur

> several orders of magnitude more than a human master encounters Not if you account for millions of years of evolution that got your nervous system to its current state and access to books and training materials which add up to multiple decades/centuries of precompiled and synthesised expertise from other people who learned chess in similar ways to you. This is a minor point though. I agree that deep learning isn't it, or that if it is, that would be quite underwhelming.

mjburgess 6y ago
Evolution produced the system, not the chess training. Humans aren't born able to play chess. The relevant comparison is "training time spent on chess".
- ad404b8a372f2b9 6y ago
  If you want to compare today's state of the art chess playing model to a human that's the relevant comparison, but if you want to compare the current field of AI to humans the line between the system and the training becomes fuzzy. If you look at the past 10 years of AI we have massively reduced the amount of data required to achieve a given score for computer vision models by making changes to the system of the neural networks. In fact it's where most of the improvements have come from rather than increasing the amount of data. I don't see a reason to believe it won't keep working this way. I'd say we're being damn efficient and it doesn't feel fair to say "look at how much data they need, the approach is fundamentally wrong" when we've been at it for such a short amount of time compared to the millions of years it took humans to evolve.
- Isinlor 6y ago
  Human designed chess, chess are made to be played by humans. If you scramble all pixels in some game in a random, but fixed way, you will never learn to play it. But speed of training of a simple feed forward network will not change at all. Here you have somewhat simpler version of what I propose: https://high-level-4.herokuapp.com/experiment https://high-level-4.herokuapp.com/experiment Here is the paper: https://rach0012.github.io/humanRL_website/ https://rach0012.github.io/humanRL_website/

mjburgess 6y ago

Evolution produced the system, not the chess training. Humans aren't born able to play chess. The relevant comparison is "training time spent on chess".

ad404b8a372f2b9 6y ago
If you want to compare today's state of the art chess playing model to a human that's the relevant comparison, but if you want to compare the current field of AI to humans the line between the system and the training becomes fuzzy. If you look at the past 10 years of AI we have massively reduced the amount of data required to achieve a given score for computer vision models by making changes to the system of the neural networks. In fact it's where most of the improvements have come from rather than increasing the amount of data. I don't see a reason to believe it won't keep working this way. I'd say we're being damn efficient and it doesn't feel fair to say "look at how much data they need, the approach is fundamentally wrong" when we've been at it for such a short amount of time compared to the millions of years it took humans to evolve.
Isinlor 6y ago
Human designed chess, chess are made to be played by humans. If you scramble all pixels in some game in a random, but fixed way, you will never learn to play it. But speed of training of a simple feed forward network will not change at all. Here you have somewhat simpler version of what I propose: https://high-level-4.herokuapp.com/experiment https://high-level-4.herokuapp.com/experiment Here is the paper: https://rach0012.github.io/humanRL_website/ https://rach0012.github.io/humanRL_website/