Live data from Hacker News

AI isn’t outthinking mathematicians, it’s out-remembering them

davidepiffer.com

451–460 of 545 posts

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#451

I suspect that a lot about what we call being very intelligent is ultimately out-remembering people around us. I think of all the times in my software career when I did something that others considered very high performance, it either came down to either having more energy than others at tackling a problem they thought was more trouble than it was worth, or just bringing back random knowledge from previous jobs or se…

>> ultimately out-remembering people around us

In order to remember something you need to know/learn it first.

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#452
post #227

I suspect that a lot about what we call being very intelligent is ultimately out-remembering people around us. I think of all the times in my software career when I did something that others considered very high performance, it either came down to either having more energy than others at tackling a problem they thought was more trouble than it was worth, or just bringing back random knowledge from previous jobs or se…

Not only I agree with you that I don't think I've had an original idea, even when I try to do artistic stuff it seems like the things most praised were the ones where I was trying to do something else and the good part came from failing to do the thing I was trying to do.

From interviews with successful creative types, that seems to be a common theme. Often, early on in their careers, they were basically trying to copy their influences, but the copy was imperfect in an interesting way.

(Also, often they're trying to copy multiple things simultaneously, which means their goal is something a bit more complex and interesting than a pure ripoff. But I think their originality really is related to the ways in which they 'fail', or, in more positive terms, the ways in which their own individuality forces its way through.)

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#453
post #220

Earlier quoted context omitted.

We sort kids into the smart and non-smart labels when they are young, and those labels tend to stick, even when what we're measuring isn't raw intelligence but just variance in childhood development that wash out in the long term. I don't think either group is well served by this.

No we don't. Public schools are incentivized not to help high performers. It actually hurts their metrics and budget. The widespread gifted/talented experience in public school ended in the late 90s. The actual intelligence segregation is self-selected. PMC parents live together in the same neighborhoods. Whether you believe in genetic or social factors, the smart kids come from here. In the 80s and 90s there was an…

I have a 1st grader. The public school did tag testing in kindergarten and then the children in the program have a special class on Mondays for the more challenging work. Is this not talented and gifted? They call it TAG.

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#454

I suspect that a lot about what we call being very intelligent is ultimately out-remembering people around us. I think of all the times in my software career when I did something that others considered very high performance, it either came down to either having more energy than others at tackling a problem they thought was more trouble than it was worth, or just bringing back random knowledge from previous jobs or se…

People have told me I was smart since I was a kid, but I can't remember for shit. I had a thought when I was fairly young that the only reason I was (maybe, sometimes) outperforming others intellectually is that I was habitually compensating for my poor memory by working things out on the fly, while others could rely more on rote memorization. Anyway, takes all kinds I guess!

Same. I often feel like my brain is a Ferrari which has a 1 L fuel tank and all the windows are blacked out except for a thumb-sized bit in the passenger side rear window.

My digit span is roughly 4 - terrible working memory. And I have trouble remembering even important things like where and how my wife and I met - terrible long term memory.

I have compensated over the years with complex systems of note taking and journals and in spite of all the above I managed a PhD, but I every day of my life feel hampered.

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#455

I suspect that a lot about what we call being very intelligent is ultimately out-remembering people around us. I think of all the times in my software career when I did something that others considered very high performance, it either came down to either having more energy than others at tackling a problem they thought was more trouble than it was worth, or just bringing back random knowledge from previous jobs or se…

asimov had a great story, "sucker bait", about this form of intelligence. (the story's premise was that we would always need some form of "human computer" because machines would never make intuitive leaps by joining facts that did not seem like they belonged together in the same problem)

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#456
post #444

Earlier quoted context omitted.

> Adding arms and 360 vision is not free. Vision is hard regardless of the angle, but scaling it from a normal camera to 360° doesn't add much cost or difficulty once you've done the hard part of turning pixels into a suitable latent space. > Look at octopi with their insane nervous system required to support their tentacle. 500 million neurons across the whole animal, brain included. We'll only know the synapse coun…

I will ignore the insane comparison between neurons and LLMs. But > This is a reason for specialists, not generalists. Precisely not. The amount of work a human can do is basically unenumerable. Requiring a specialist for each task will quickly run up costs to be completely unreasonable.

> I will ignore the insane comparison between neurons and LLMs. But

Please don't.

I mean, you yourself specifically said to "Look at octopi with their insane nervous system", and that's exactly what I'm doing and my conclusion is "doesn't look that bad compared to what else we do now".

Also note that I expressly noted that synapses, not neurons, are the important factor in parameter count.

> The amount of work a human can do is basically unenumerable.

https://www.bls.gov/ooh/a-z-index.htm

> Requiring a specialist for each task will quickly run up costs to be completely unreasonable.

The existence of specialists in professional settings proves this claim false.

As does the existence of special-purpose robots, all the way down to injection moulds designed for one specific part, which overall make the end product cheaper, a process which probably started with whoever invented whichever came first of the wind or water powered grain mill.

Much of the economic gain of the industrial revolution was replacing the very expensive salaries of general-purpose humans with machines that, despite being extremely single-purpose and having high purchase and maintenance costs, were still cheaper than the wages of the huge number of humans needed for equal output.

Since the industrial revolution we have been the species that makes tools even though it sometimes also uses slavery, not the species that relies on slaves and has tools for decorative reasons.

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#458
post #328

I suspect that a lot about what we call being very intelligent is ultimately out-remembering people around us. I think of all the times in my software career when I did something that others considered very high performance, it either came down to either having more energy than others at tackling a problem they thought was more trouble than it was worth, or just bringing back random knowledge from previous jobs or se…

> out-remembering people around us. Doing an angular -> react refactor with Opus 5 at work. Amazed at how good it is but also amazed at how dumb it is after I told it I want a very basic and specific code review:do the file names match the contents? is one functionality/concern split in multiple files? do the variable names match what they do? does a method read nicely? Anyway, I was very surprised at how bad it can…

> I suspect it's because the LLM simply does not focus well enough on a single method.

Consider asking it to use sub-agents or dynamic workflows for that, then the context will be polluted less by the surrounding stuff, UNLESS you wanna babysit it through let's say 5-10 files first and then write down the desired results in some plan file that each of the sub-agents would get.

Personally, Opus 5 has been kinda inconsistent for me - sometimes the technical capabilities are very much there, other times it's a bit dumb, but almost always the tone is insufferable and obnoxious LinkedIn speak, no matter what I try. Nowadays using Kimi K3 and GLM 5.3 more, though they exhibit similar issues with performance, grass is greener only in regards to tone.

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#459

I suspect that a lot about what we call being very intelligent is ultimately out-remembering people around us. I think of all the times in my software career when I did something that others considered very high performance, it either came down to either having more energy than others at tackling a problem they thought was more trouble than it was worth, or just bringing back random knowledge from previous jobs or se…

People have told me I was smart since I was a kid, but I can't remember for shit. I had a thought when I was fairly young that the only reason I was (maybe, sometimes) outperforming others intellectually is that I was habitually compensating for my poor memory by working things out on the fly, while others could rely more on rote memorization. Anyway, takes all kinds I guess!

"Working out on the fly" - I do the same thing, I try to find the minimum I can get away with memory-wise, and try to connect everything else to that in a way I can retrace my steps easily. When I was in school/uni, my notes started out as say 10 pages per lecture, and got thinned down over the months until I only had one page or half a page to memorize on exam day, and everything else could be decompressed from that nugget.

Re: AI isn’t outthinking mathematicians, it’s out-remembering them

#460

That intelligence is "just" retrieval within a huge corpus is an old one, and the one I am thinking about a lot these days: https://news.ycombinator.com/item?id=44060672

I think there's definitely more than just retrieval, although no doubt it helps.
Post reply on HN