• 5 Posts
  • 272 Comments
Joined 9 months ago
cake
Cake day: March 2nd, 2024

help-circle


  • As an egalitarian, I also inherently dislike divisions in competition based on demographic. Here’s my understanding of why they exist (though my stating the justification here doesn’t mean I tacitly agree):

    Chess: currently, only about 40 of the 1600 grandmasters are women. To attain a balance, we ought to be encouraging women to play chess. Women-only competitions are a great way to do this. (There are almost no transgender chess grand/masters, so the same logic ought to apply here – I don’t understand any reason other than bigotry to exclude transgender women from such tournaments)

    Sports: I think it comes down to a Schelling division. Now sure, there are other genetic advantages, perhaps race or leg length or height or other aspects influence one’s athletic ability too – top basketball players are generally many standard deviations above average height. However, those are spectra – ranges – so there’s no obvious place to split into two categories. There are basically only two obvious, bright-line, ostensibly binary dichotomies that people tend to believe categorize humans: (a) sex, and (b) disabled status (see: paralympics).

    Now, imagine there was a genetic allele that causes humans to be 9 feet tall. About half of humans get this allele. Then obviously we’d add a new category for these super-tall humans, just so that less-tall humans would have the option to compete in sports.

    Some sports make divisions on a spectrum, like heavy-weight, medium-weight, light-weight boxing and so on. But these are pretty arbitrary, certainly not Schelling points, so it’s less common for sports to use these divisions.

    Now, I often find myself thinking, shouldn’t those certain cis men who happen by nature to be less able than a typical woman be permitted in the women’s category? My gut answer is yes – but the problem here is that there’s just no way to measure someone’s natural capacity for ability. There’s no bright-line, Schelling-point way to sort out these less-capable cis men. It sucks.


  • Exactly. So I am curious why OP considers it remarkable that “neither left nor right” is not what it seems.

    As for me, I might consider someone from a different country with different politics, like Japan perhaps, to be neither. Or someone who lives under a rock and doesn’t pay attention to the news.
















  • LLMs are basically just good pattern matchers. But just like how A* search can find a better path than a human can by breaking the problem down into simple steps, so too can an LLM make progress on an unsolved problem if it’s used properly and combined with a formal reasoning engine.

    I’m going to be real with you: the big insight behind almost all new mathematical ideas is based on the math that came before. Nothing is truly original the way AI detractors seem to believe.

    By “does some reasoning steps,” OpenAI presumably are just invoking the LLM iteratively so that it can review its own output before providing a final answer. It’s not a new idea.


  • I do agree that grad students don’t exactly live in luxury, and frequently develop mental health crises. But their contributions and insight are what power their labs. Profs often have to spend so much time teaching and chasing grants that they can’t do much real research. Academia overall is in a sad state.

    But Tao is a superstar, and a charismatic blogger. I’d be disappointed to learn he mistreats his grad students. (I don’t know if he even has any tbh)