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

A non-technical explanation of deep learning

parand.com

Post reply on HN
Print view View original

131–139 of 139 posts

Re: A non-technical explanation of deep learning

#131

Post by dllthomas »

robotresearcher wrote: post #117
dragonwriter wrote: post #99

Earlier quoted context omitted.

A progressive income tax that does not exclude or favor capital income, so that funding ideally targeted transitional assistance, UBI (with a rate ratcheting up with sustained increases in per capita revenue), or, ideally, both so that the adverse effects of labor market shifts which shuffle or concentrate labor demand or shift from labor-intensive to capital-intensive methods are buffered.

How do you prevent UBI being captured by landlords in rent, and similar non-discretionary spending?

Competition, both between similar units in similar places and between alternative arrangements. Maybe having a roommate or living with my parents isn't worth it for $X/mo but it is for $2X. Certainly it is for some X for most people. Also the presence of a UBI directly increases "housing" in the sense we mean. We don't really have any shortage of housing - we have a shortage of attractive housing, where a huge part of "attractive" is "sufficiently close to sufficient income" and a UBI makes more things sufficient. And for those in the cities, by elevating the bottom of the income distribution we make developers more interested in building the affordable housing we need.

Will some of the increase still be captured? Probably. More in some areas than others. Will we be right back to square one (or worse)? I really don't think so.

Re: A non-technical explanation of deep learning

#132

Post by ftxbro »

amelius wrote: post #124
ftxbro wrote: post #63

Earlier quoted context omitted.

> Advanced math is mostly useless because of the dimensionality of neural nets. It depends what you mean by advanced math. There is a lot of math that only really comes into play because of the high dimensionality! For example math related to tensor wrangling, low rank approximations, spectral theory, harmonic theory, matrix calculus derivatives, universality principles, and other concepts that could be interesting o…

Yes but do you have examples of "higher" math not being just a curiosity and actually making it into real world models and training algorithms?

Well I suppose that in some sense you are right. You can do deep learning without even knowing any math at all, by plugging together libraries and frameworks that other people wrote.

Also maybe you will say that "higher" math is by definition a curiosity and if it's practical then it's not "higher".

But if those aren't your arguments, then you can consider one example that the tensor 'differentiable programming' libraries used in deep learning use automatic differentiation and matrix calculus. Matrices are taught in high school, and calculus is taught in high school, but matrix calculus generally isn't as far as I know. Or at least not at my high school. https://en.wikipedia.org/wiki/Matrix_calculus

Re: A non-technical explanation of deep learning

#133

Post by musicale »

ketzo wrote: post #91
robotresearcher wrote: post #90

Earlier quoted context omitted.

Another part of the answer over the last 40 (or 200) years, is to repeatedly create totally new industries that employ lots of people, including a large fraction of HN readers.

Yeah, but what does the Venn diagram look like there? Yes, new technologies create new jobs. But it’s not usually the people from the old jobs who are taking those new jobs.

> it’s not usually the people from the old jobs who are taking those new jobs

https://news.mit.edu/2022/automation-drives-income-inequalit...

Re: A non-technical explanation of deep learning

#134

Post by musicale »

robotresearcher wrote: post #117
dragonwriter wrote: post #99

Earlier quoted context omitted.

A progressive income tax that does not exclude or favor capital income, so that funding ideally targeted transitional assistance, UBI (with a rate ratcheting up with sustained increases in per capita revenue), or, ideally, both so that the adverse effects of labor market shifts which shuffle or concentrate labor demand or shift from labor-intensive to capital-intensive methods are buffered.

How do you prevent UBI being captured by landlords in rent, and similar non-discretionary spending?

In the current US system, much of income is captured by rent and health insurance, regardless of where it comes from.

But I share the same concern that UBI would immediately be added to rent.

And that it might combine poorly with existing perverse incentives:

"Across the nation, landlords with units in poor neighborhoods average nearly $100 a month in net profit, compared to about $50 in affluent neighborhoods, and just $3 in middle-class areas."

https://www.bloomberg.com/news/articles/2019-03-21/housing-e...

Re: A non-technical explanation of deep learning

#135

Post by cscurmudgeon »

wrs wrote: post #80
cscurmudgeon wrote: post #56

Earlier quoted context omitted.

Thats not the Chinese Room argument. The argument says just because a system processes X doesn't imply it has consciousness of X.

The flaw is the unsupported assertion that the whole system being conscious of X depends on a part of the system being conscious of X. The same assertion would fail here in the same way.

Are you talking about access consciousness or phenomenal consciousness?

Re: A non-technical explanation of deep learning

#136

Post by musicale »

dllthomas wrote: post #131
robotresearcher wrote: post #117

Earlier quoted context omitted.

How do you prevent UBI being captured by landlords in rent, and similar non-discretionary spending?

Competition, both between similar units in similar places and between alternative arrangements. Maybe having a roommate or living with my parents isn't worth it for $X/mo but it is for $2X. Certainly it is for some X for most people. Also the presence of a UBI directly increases "housing" in the sense we mean. We don't really have any shortage of housing - we have a shortage of attractive housing, where a huge part o…

That is an interesting argument: increasing rent might make it financially attractive to move in with roommates or family even if you could afford the rental.

I'm not sure that's always an option if you have to live near your workplace, but it could shake things up a bit.

Re: A non-technical explanation of deep learning

#137

Post by dllthomas »

musicale wrote: post #136
dllthomas wrote: post #131

Earlier quoted context omitted.

Competition, both between similar units in similar places and between alternative arrangements. Maybe having a roommate or living with my parents isn't worth it for $X/mo but it is for $2X. Certainly it is for some X for most people. Also the presence of a UBI directly increases "housing" in the sense we mean. We don't really have any shortage of housing - we have a shortage of attractive housing, where a huge part o…

That is an interesting argument: increasing rent might make it financially attractive to move in with roommates or family even if you could afford the rental. I'm not sure that's always an option if you have to live near your workplace, but it could shake things up a bit.

Yeah, it's certainly not always an option, but it just needs to be enough of an option for enough people in order to put some downward pressure on rents (compared to full capture of the increased income).

Re: A non-technical explanation of deep learning

#138

Post by melagonster »

dmreedy wrote: post #87
commandlinefan wrote: post #61

Earlier quoted context omitted.

> an appropriate level of understanding mainly through analogies and metaphors I think it's actually worse than that - somebody who doesn't know actually realizes that he doesn't know, but somebody who _thinks_ he understands through analogies and metaphors will confidently come to the incorrect conclusion and then argue with somebody who actually does understand the topic - often managing to convince innocent bystan…

I am fascinated by this phenomenon, and the double-edged sword that metaphors are. On the one hand they're jargon used as short hand to technical concepts understood well by domain experts. And the concision they afford can lead to deeper understanding as they transcend their composite or adapted meanings and become base terminology in and of themselves (I think of e.g. Latin in English legal terminology. "Habeas cor…

this is how we teach students in all discipline. displaying a model and then offer better one, again and again.

Re: A non-technical explanation of deep learning

#139

Post by wrs »

pringk02 wrote: post #122
tasty_freeze wrote: post #28

Earlier quoted context omitted.

> the Chinese Room argument is only arguing against the much narrower, and now philosophically unfashionable, "Hard AI" stance as it was held in the 70s Searle has stood behind his argument in the 70s, but in every decade since then too. The main failure is that most people fundamentally don't believe they are mechanistic. If one believe in dualism, then it easy to attribute various mental states to that dualism, and…

Searle's argument in the Chinese Room is only that passing the Turing Test isn't enough to prove evidence of Mind (capital 'm' to distinguish it at the philosophical jargon term, and all it entails). He does hold the stance that he doesn't think Computationalism (in the style of Dennet) is correct. I'm not sure if he personally feels the Chinese Room argument refutes that stance as a whole, but I believe the general…

I find the argument to be essentially assuming the premise (if a system acts conscious, but you look inside it and don’t find anything conscious in there, then it can’t be conscious) but honestly I wouldn’t say I’m sure I understand the argument. Given my understanding, I thought this illustrated the circularity. But I see I’m wrong, because if you buy the premise then you won’t find the situation analogous.

That is, imagine there was a famous philosopher who insisted that cats can only be recognized by some non-computational mechanism, and any computational cat recognizer might “simulate” recognizing cats but could not be said to actually recognize cats. Then you build a neural network that recognizes cats, and they open it up and point out that nothing in it can recognize cats, so therefore it isn’t “really” recognizing cats.

Post reply on HN