Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Fitting an elephant with four non-zero parameters

arxiv.org

Post reply on HN
Print view View original

51–60 of 155 posts

Re: Fitting an elephant with four non-zero parameters

#51

Post by tobias2014 »

edflsafoiewq wrote: post #48
maitola wrote: post #16

I love the ironic side of the article. Perhaps they should add the reason for it, from Fermi's and Neumann's. When you are building a model of reality in Physics, If something doesn’t fit the experiments, you can’t just add a parameter (or more) variate it and fit the data. The model should have zero parameters, ideally, or the least possible, or, even at a more deeper level, the parameters should emerge naturally fr…

Isn't the form of an equation really just another sort of parameter?

This is why I think that modeling elementary physics is nothing else than fitting data. We might end up with something that we perceive as "simple", or not. But in any case all the fitting has been hidden in the process of ruling out models. It's just that a lot of the fitting process is (implicitly) being done by theorists; we come up with new models and that are then being falsified.

For example, how many parameters does the Standard Model have? It's not clear what you count as a parameter. Do you count the group structure, the other mathematical structure that has been "fitted" through decades of comparisons with experiments?

Re: Fitting an elephant with four non-zero parameters

#52

Post by gjm11 »

mananaysiempre wrote: post #22
btown wrote: post #6

Earlier quoted context omitted.

There's plenty of humor on arXiv, and that's part of why it's so incredible! Some lists: https://academia.stackexchange.com/questions/86346/is-it-ok-... https://www.ellipsix.net/arxiv-joke-papers.html

Joke titles and/or author lists are also quite popular, e.g. the Greenberg, Greenberger, Greenbergest paper[1], a paper with a cat coauthor whose title I can’t seem to recall (but I’m sure there’s more than one I’ve encountered), or even the venerable, unfortunate in its joke but foundational in its substance Alpher, Bethe, Gamow paper[2]. Somewhat closer to home, I think computer scientist Conor McBride[3] is the ch…

> paper with a cat coauthor whose title I can't seem to recall

You probably have in mind https://en.wikipedia.org/wiki/F._D._C._Willard (coauthor of multiple papers, sole author of at least one).

Re: Fitting an elephant with four non-zero parameters

#55

Post by ahazred8ta »

maitola wrote: post #16

I love the ironic side of the article. Perhaps they should add the reason for it, from Fermi's and Neumann's. When you are building a model of reality in Physics, If something doesn’t fit the experiments, you can’t just add a parameter (or more) variate it and fit the data. The model should have zero parameters, ideally, or the least possible, or, even at a more deeper level, the parameters should emerge naturally fr…

Notably done for the first time irl in "Least square fitting of an elephant", James Wei (1975) Chemtech

Re: Fitting an elephant with four non-zero parameters

#56

Post by rocqua »

karmakaze wrote: post #43
jampekka wrote: post #21

Earlier quoted context omitted.

This was mentioned in the first paragraph of the paper. The paper is mostly humoristic. That said, the wisdom of the quip has been widely lost in many fields. In many fields data is "modeled" with huge regression models with dozens of parameters or even neural networks with billions of parameters. > In 1953, Enrico Fermi criticized Dyson’s model by quoting Johnny von Neumann: “With four parameters I can fit an elepha…

That's how I feel about dark matter. Oh this galaxy is slower than this other similar one. The first one must have less dark matter then. What can't be fit by declaring the amount of dark matter that must be present fits the data? It's unfalsifiable, just because we haven't found it, doesn't mean it doesn't exist. Even worse than string/M-theory which at least has math.

I used to think this, but dark matter does make useful predictions, that are hard to explain otherwise.

This is partially because there are two ways to detect dark-matter. The first is gravitational lensing. The second is the rotatinal speed of galaxies. There are some galaxies that need less Dark Matter to explain their rotational speed. We can then cross check whether those galaxies cause less gravitational lensing.

Besides that, the gravitational lensing of galaxies being stronger than the bright matter in the galaxies can justify is hard to explain without dark matter.

Re: Fitting an elephant with four non-zero parameters

#57

Post by rocqua »

edflsafoiewq wrote: post #48
maitola wrote: post #16

I love the ironic side of the article. Perhaps they should add the reason for it, from Fermi's and Neumann's. When you are building a model of reality in Physics, If something doesn’t fit the experiments, you can’t just add a parameter (or more) variate it and fit the data. The model should have zero parameters, ideally, or the least possible, or, even at a more deeper level, the parameters should emerge naturally fr…

Isn't the form of an equation really just another sort of parameter?

It tends to be a parameter that can be derived from rrasoning and assumptions. This contrasts to free parameters where you say "and we have no idea what this value should be, so we'll measure it"

Re: Fitting an elephant with four non-zero parameters

#58

Post by jampekka »

rocqua wrote: post #56
karmakaze wrote: post #43

Earlier quoted context omitted.

That's how I feel about dark matter. Oh this galaxy is slower than this other similar one. The first one must have less dark matter then. What can't be fit by declaring the amount of dark matter that must be present fits the data? It's unfalsifiable, just because we haven't found it, doesn't mean it doesn't exist. Even worse than string/M-theory which at least has math.

I used to think this, but dark matter does make useful predictions, that are hard to explain otherwise. This is partially because there are two ways to detect dark-matter. The first is gravitational lensing. The second is the rotatinal speed of galaxies. There are some galaxies that need less Dark Matter to explain their rotational speed. We can then cross check whether those galaxies cause less gravitational lensing…

The problem with dark matter is that there's no (working) theory on how the dark matter is distributed. It's really easy to "explain" gravitational effects if you can postulate extra mass ad-hoc to fit the observations.

Re: Fitting an elephant with four non-zero parameters

#59

Post by aqme28 »

Sesse__ wrote: post #46
aqme28 wrote: post #34

> It only satisfies a weaker condition, i.e., using four non-zero parameters instead of four parameters. Why would that be a harder problem? In the case that you get a zero parameter, you could inflate it by some epsilon and the solution would basically be the same.

They also, effectively, fit information in the indexes of the parameters. I.e., _which_ of the parameters are nonzero carries real information. In a sense, they have done their fitting using nine parameters, of which five are zero.

I didn’t read enough to catch that. How the heck did they justify that?

Re: Fitting an elephant with four non-zero parameters

#60

Post by rocqua »

soist wrote: post #18
visarga wrote: post #15

Earlier quoted context omitted.

> Whether "just curve fitting" is enough to produce something "intelligent" is, IMO, currently unanswered Continual "curve fitting" to the real world can create intelligence. What is missing is not something inside the model. It's missing a mechanism to explore, search and expand its experience. Our current crop of LLMs ride on human experience, they have not largely participated in creating their own experiences. Th…

AlphaZero did not create any experiences. AlphaZero was software written by people to play board games and that's all it ever did.

Are you launching into a semantic argument about the word 'experience'? If so, it might help to state what essential properties alphago was missing that makes it 'not having an experience'.

Otherwise this can quickly devolve into the common useless semantic discussion.

Post reply on HN