3 ms·

“What the I.M.O. is testing is very different from what creative mathematics looks like for the vast majority of mathematicians,” he said. --- Not to pick on t

“What the I.M.O. is testing is very different from what creative mathematics looks like for the vast majority of mathematicians,” he said. --- Not to pick on this guy, but this is ridiculous goal post shifting. It's just astounding what people will hand-wave away as not requiring intelligence.

mb7733 3y ago
No, that sort of thing has been said about math competitions for a long time. It's not a new argument put forward as something against AI. An analogy with software is that math competitions are like (very hard) leetcode. There was an article posted on HN recently that is related: https://benexdict.io/p/math-team https://benexdict.io/p/math-team
- esafak 3y ago
  What is the argument; that math competitions are easy for computers but hard for humans?
  - nitwit005 3y ago
    It's a quote from the article. The argument is naturally there.
    - esafak 3y ago
      I was referring to your leetcode analogy; those too are hard for humans.
      - nyssos 3y ago
        No, the analogue of competition math here is writing programs to solve leetcode problems: they emphasize quickly and reliably applying known tools, not developing new ones.
  - silveraxe93 3y ago
    The usual argument is: We test developer skill by giving them leetcode problems, but leetcode while requiring programming skill is nothing like a real programmer's job.
  - mb7733 3y ago
    Like I said, it's got nothing to do with computers or AI. This point of view predates any kind of AI that would be capable of doing either. The analogy is as follows. Like with with leetcode & job interviews is as follows, to excel at math competitions one must grind problems and learn tricks, in order to quickly solve problems in a high pressure environment. And, just like how solving leetcode problems is pretty different than what a typical computer scientist or software engineer does, doing math competitions is pretty different than what a typical mathematician does.
drawnwren 3y ago
Yeah, this would be akin to saying "What leetcode is testing is very different from what professional programming looks like". It isn't untrue, but both are considered metrics by the community. (Whether rightfully or not seems irrelevant to ML).
adverbly 3y ago
I guess it kind of depends on what a goal post should represent. If it represents an incremental goal then obviously you are correct. I think the ultimate goal though is clearly that AI systems will be able to generate new high-value knowledge and proofs. I think the realistically the final goal post has always been at that point.
GPerson 3y ago
I’m not sure if it’s goal post shifting or not, but it is a true statement.
nybsjytm 3y ago
As said by others, that is an absolutely commonplace saying, albeit usually having nothing whatsoever to do with AI. See for instance https://terrytao.wordpress.com/career-advice/advice-on-mathematics-competitions/ https://terrytao.wordpress.com/career-advice/advice-on-mathe... Also if you read about the structure of AlphaGeometry, I think it's very hard to maintain that it "requires intelligence." As AI stuff goes, it's pretty comprehensible and plain.

mb7733 3y ago

No, that sort of thing has been said about math competitions for a long time. It's not a new argument put forward as something against AI. An analogy with software is that math competitions are like (very hard) leetcode. There was an article posted on HN recently that is related: https://benexdict.io/p/math-team https://benexdict.io/p/math-team

esafak 3y ago
What is the argument; that math competitions are easy for computers but hard for humans?
- nitwit005 3y ago
  It's a quote from the article. The argument is naturally there.
  - esafak 3y ago
    I was referring to your leetcode analogy; those too are hard for humans.
    - nyssos 3y ago
      No, the analogue of competition math here is writing programs to solve leetcode problems: they emphasize quickly and reliably applying known tools, not developing new ones.
- silveraxe93 3y ago
  The usual argument is: We test developer skill by giving them leetcode problems, but leetcode while requiring programming skill is nothing like a real programmer's job.
- mb7733 3y ago
  Like I said, it's got nothing to do with computers or AI. This point of view predates any kind of AI that would be capable of doing either. The analogy is as follows. Like with with leetcode & job interviews is as follows, to excel at math competitions one must grind problems and learn tricks, in order to quickly solve problems in a high pressure environment. And, just like how solving leetcode problems is pretty different than what a typical computer scientist or software engineer does, doing math competitions is pretty different than what a typical mathematician does.

drawnwren 3y ago

Yeah, this would be akin to saying "What leetcode is testing is very different from what professional programming looks like". It isn't untrue, but both are considered metrics by the community. (Whether rightfully or not seems irrelevant to ML).

adverbly 3y ago

I guess it kind of depends on what a goal post should represent. If it represents an incremental goal then obviously you are correct. I think the ultimate goal though is clearly that AI systems will be able to generate new high-value knowledge and proofs. I think the realistically the final goal post has always been at that point.

GPerson 3y ago

I’m not sure if it’s goal post shifting or not, but it is a true statement.

nybsjytm 3y ago

As said by others, that is an absolutely commonplace saying, albeit usually having nothing whatsoever to do with AI. See for instance https://terrytao.wordpress.com/career-advice/advice-on-mathematics-competitions/ https://terrytao.wordpress.com/career-advice/advice-on-mathe... Also if you read about the structure of AlphaGeometry, I think it's very hard to maintain that it "requires intelligence." As AI stuff goes, it's pretty comprehensible and plain.