3 ms·
> For example, humans are much better at learning languages compared to even simple mathematics. Even those who are "good at math" are objectively much better a
by bee_rider 6y ago
> For example, humans are much better at learning languages compared to even simple mathematics. Even those who are "good at math" are objectively much better at languages, in terms of how much complexity they can absorb.
Is this actually the general consensus? I think this is wrong. I am mediocre at math, but I can write it, and then verify that it is correct, and then it is quite abstractable and transferrable. Conventional languages handle fuzzy input better, but they do that by allowing ambiguous parsing..
> A14: Deterministic beats heuristic.
> Heuristics make it hard to give precise safety guarantees, so you should only use them when you don't guarantee much.
I think this is overly broad. If you can run a heuristic and then validate the results in a way that provides a guarantee, that can often be just as good, right? You've still got the bound, you just had to wait until after the computation to get it. Sometimes you'll get a failure to validate instead, but that should be rare if the heuristic is good. Anyway, all our programs are relying on the heuristic "hardware usually works."
I mean, nobody loves heuristics, but we keep coming up with them, so that must be a sign, right?
- jerf 6y ago"I am mediocre at math, but I can write it, and then verify that it is correct, and then it is quite abstractable and transferrable." Imagine you encountered a meaningful equation with as many symbols as you have in that sentence. Count characters, or even count words+punctuation if you like. Odds are, you are not going to apprehend that equation with anything like the fluidity or speed your wrote your sentence or I read it... yet if you really dig into it, such as by diagramming, your sentence is a quite non-trivial expression. That you are handling fuzzy parsing, and doing so so well, argues for you being better at language understanding than mathematics rather than against it. Trying to read fuzzily-serialized math is even worse. The average IQ-100 human being will on-the-fly both generate and comprehend in real time fairly complicated grammar structures, even with no formal training, while at the same time, it takes years of formal training to even begin to process very simple mathematical expressions. On the flip side, one can argue that there are other considerations. What we call "math" has a very different shape that the things we use language to describe... "sharper", all the details matter, etc. On the other hand, my day job, like most of us here, is slinging around many thousands of mathematical constructs fairly casually as well, more like a language than doing math. I'm using fairly mathematical tools, but I often have the impression that I'm spraying them around like some sort of quick-setting concrete coming out of a hose, rather than crafting magnificent cathedrals of precisely-chosen abstractions, and there's a lot of reasons to prefer the former to the latter. On the flip flip side, getting to the point that I can do this took me longer than it took to speak in English, and it doesn't appear to be something the average human can do very well.... Basically, the upshot of my point here is that language users casually manipulate hundreds of symbols spontaneously so easily that it barely even registers as effort to you, whereas math is generally not so easy. And even if the "shape" of language is somehow easier, it's not like it isn't precise. This sentance is incorrect, despite only a small error. This incorrect is sentence too. Math may be more precise than language, but it's not like it's a night-and-day contrast; if you look at language carefully enough you can see it's still fairly high on the precision scale overall.