When to assume neural networks can solve a problem • Hacker News Board

Re: When to assume neural networks can solve a problem

#11

Post by gambler » Mon, Mar 30, 2020, 2:53 PM UTC

forgotmyhnacc wrote: post #9

gambler wrote: post #8

I've been hearing about deep learning revolutionizing "everything" for the past 8 years. So, can someone name me any significant impact it made on figuring out what to do about the recent epidemic? If not, I think it's worth reflecting on what value we get out of the technology that sucked up so much of our intellectual, financial and computational resources.

I've been hearing about modern JavaScript development revolutionizing "everything" for the past 8 years. So, can someone name me any significant impact it made on figuring out what to do about the recent epidemic? If not, I think it's worth reflecting on what value we get out of the technology that sucked up so much of our intellectual, financial and computational resources.

This is a nonsensical response on two levels.

First, you're missing the obvious fact that the response to this pandemic is largely dictated by data processing. People are modeling the spread of the virus, trying to make predictions about the optimal policy, trying to figure out infection rates without full testing and evaluating promising drugs. All of those things relate to modeling and data analysis. Deep learning was touted to revolutionize exactly those two things.

Second, web technologies are making an obvious impact by allowing (some, many) people to stay in touch, stay informed, keep working and buy stuff without exposing themselves to crowded places. This constitutes visible and significant positive impact.

Re: When to assume neural networks can solve a problem

#12

Post by _f9cu » Mon, Mar 30, 2020, 2:53 PM UTC

When a neural network solved a problem, it means it found the algebraic function it needs to solve the problem.

> Once it found the algebraic function, there is no need to run the problem to the neural network. We can sideload the problem to a simple program that takes input and gives an output. This strategy can help to free up the GPU for next set of problem.

Re: When to assume neural networks can solve a problem

#13

Post by tmoravec » Mon, Mar 30, 2020, 2:59 PM UTC

kelvin0 wrote: post #10

makach wrote: post #3

This reminds me of an oldie, but goodie: http://thedailywtf.com/articles/No%2C_We_Need_a_Neural_Netwo...

Wow, I had never read this gem before. Made me laugh, thanks for sharing this piece! I'm surprised they did not ask him to log everything in a blockchain :)

I think that's only because the story is from 2006 so blockchain was not a thing, yet.

Re: When to assume neural networks can solve a problem

#14

Post by teraflop » Mon, Mar 30, 2020, 2:59 PM UTC

kelvin0 wrote: post #10

makach wrote: post #3

This reminds me of an oldie, but goodie: http://thedailywtf.com/articles/No%2C_We_Need_a_Neural_Netwo...

Wow, I had never read this gem before. Made me laugh, thanks for sharing this piece! I'm surprised they did not ask him to log everything in a blockchain :)

Check the date -- the article predates the Bitcoin white paper by two years. (It also predates most of the modern "deep learning" hype.)

Re: When to assume neural networks can solve a problem

#15

Post by folbec » Mon, Mar 30, 2020, 3:20 PM UTC

"That’s why building a product recommendation algorithm was a hot topic 20 years ago, but nowadays everyone and their mom can just get a WordPress plugin for it and get close to Amazon’s level."

Well, I haven't had a remotely relevant Amazon recommendation in 5 to 10 years. Unless you count read a book by the exact same author as a useful recommendation.

Re: When to assume neural networks can solve a problem

#16

Post by ludwigschubert » Mon, Mar 30, 2020, 3:42 PM UTC

_f9cu wrote: post #12

When a neural network solved a problem, it means it found the algebraic function it needs to solve the problem. > Once it found the algebraic function, there is no need to run the problem to the neural network. We can sideload the problem to a simple program that takes input and gives an output. This strategy can help to free up the GPU for next set of problem.

How would one extract an algebraic function from the network’s weights? Why would the function the network approximates necessarily be algebraic?

Re: When to assume neural networks can solve a problem

#17

Post by scrooched_moose » Mon, Mar 30, 2020, 4:11 PM UTC

folbec wrote: post #15

"That’s why building a product recommendation algorithm was a hot topic 20 years ago, but nowadays everyone and their mom can just get a WordPress plugin for it and get close to Amazon’s level." Well, I haven't had a remotely relevant Amazon recommendation in 5 to 10 years. Unless you count read a book by the exact same author as a useful recommendation.

At least on my account, it seems like Amazon has given up on recommendations and is just suggesting I buy the same things again.

I wouldn't be shocked if it's actually quite a bit more accurate then real predictions, but it does lead to funny things like "You already bought this book? How about a second copy?"

Re: When to assume neural networks can solve a problem

#18

Post by malux85 » Mon, Mar 30, 2020, 4:33 PM UTC

gambler wrote: post #8

I've been hearing about deep learning revolutionizing "everything" for the past 8 years. So, can someone name me any significant impact it made on figuring out what to do about the recent epidemic? If not, I think it's worth reflecting on what value we get out of the technology that sucked up so much of our intellectual, financial and computational resources.

If you're listening to people who claim any one technology solves "everything" --- and believing them --- then you are the fool.

There's no silver bullet. Deep Learning has made huge strides in noise reduction, image and video processing (self driving cars), lidar processing, medical imaging, etc, but it's not made any progress in knitting, tooth brushing, dog walking, or many other tasks ill suited for deep learning - so clearly not everything.

Re: When to assume neural networks can solve a problem

#19

Post by yters » Mon, Mar 30, 2020, 4:38 PM UTC

malux85 wrote: post #18

gambler wrote: post #8

I've been hearing about deep learning revolutionizing "everything" for the past 8 years. So, can someone name me any significant impact it made on figuring out what to do about the recent epidemic? If not, I think it's worth reflecting on what value we get out of the technology that sucked up so much of our intellectual, financial and computational resources.

If you're listening to people who claim any one technology solves "everything" --- and believing them --- then you are the fool. There's no silver bullet. Deep Learning has made huge strides in noise reduction, image and video processing (self driving cars), lidar processing, medical imaging, etc, but it's not made any progress in knitting, tooth brushing, dog walking, or many other tasks ill suited for deep learning…

That's been the general vibe the popular tech press has been spouting for awhile. That NNs are somehow a major step towards AGI, which'll 'capture the lightcone of the future' to quote Sam Altman.

At some point, we'll realize that AGI is the modern alchemy. Ironically, alchemy's penultimate goal was to construct a homoculus, which is essentially the goal of AGI.

https://en.wikipedia.org/wiki/Homunculus#Alchemy

"The allegorical text suggests to the reader that the ultimate goal of alchemy is not chrysopoeia, but it is instead the artificial generation of humans."

Re: When to assume neural networks can solve a problem

#20

Post by malux85 » Mon, Mar 30, 2020, 4:44 PM UTC

yters wrote: post #19

malux85 wrote: post #18

Earlier quoted context omitted.

If you're listening to people who claim any one technology solves "everything" --- and believing them --- then you are the fool. There's no silver bullet. Deep Learning has made huge strides in noise reduction, image and video processing (self driving cars), lidar processing, medical imaging, etc, but it's not made any progress in knitting, tooth brushing, dog walking, or many other tasks ill suited for deep learning…

That's been the general vibe the popular tech press has been spouting for awhile. That NNs are somehow a major step towards AGI, which'll 'capture the lightcone of the future' to quote Sam Altman. At some point, we'll realize that AGI is the modern alchemy. Ironically, alchemy's penultimate goal was to construct a homoculus, which is essentially the goal of AGI. https://en.wikipedia.org/wiki/Homunculus#Alchemy "The a…

> That's been the general vibe the popular tech press has been spouting for awhile

This is such a vague generalisation, I cant do anything with it, except maybe how about "nullius in verba", on the word of noone, or "think for yourself"