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

Evolution Is the New Deep Learning

sentient.ai

Post reply on HN
Print view View original

191–200 of 242 posts

Re: Evolution Is the New Deep Learning

#191

Post by dorgo »

goatlover wrote: post #174
dorgo wrote: post #120

Earlier quoted context omitted.

> My gut feeling is that it is hubris to think that we are going to "figure out" intelligence with increasingly sophisticated mathematical models anytime soon. We did it already. Compter understand language, translate it, react to it. They can recognize items on a picture. Is there a task left which can't be done by computers better and faster than by humans? >Almost by definition, if we can analytically understand i…

> Is there a task left which can't be done by computers better and faster than by humans? All the tasks humans still earn money doing. And given that we're nowhere near full automation, I'd say it's quite a few tasks.

Like the collector in the supermarket who scans the products and takes my cash? Surely it must be impossible to automate such a complex task.

Industry/Economy lacks behind state-of-the-art technology by decades.

Re: Evolution Is the New Deep Learning

#192

Post by dorgo »

jcora wrote: post #154
dorgo wrote: post #120

Earlier quoted context omitted.

> My gut feeling is that it is hubris to think that we are going to "figure out" intelligence with increasingly sophisticated mathematical models anytime soon. We did it already. Compter understand language, translate it, react to it. They can recognize items on a picture. Is there a task left which can't be done by computers better and faster than by humans? >Almost by definition, if we can analytically understand i…

You should study ML a bit more, to see just how wrong you are about it

I said two things. Which am I wrong about?

Re: Evolution Is the New Deep Learning

#193

Post by dorgo »

oldandtired wrote: post #170
tree_of_item wrote: post #121

Earlier quoted context omitted.

> Is there a task left which can't be done by computers better and faster than by humans? Are you serious? You think we're done?

Probably better if you gave a task example like being a mother or a father or a grandparent or an uncle or an aunt or a mentor or a friend or anything to do with human interaction.

There is no market for fathers and uncles. Nobody wants a roboter as father. Mentor of friend sounds more interesting though.

Re: Evolution Is the New Deep Learning

#194

Post by yters »

_mhr_ wrote: post #188
yters wrote: post #123

Earlier quoted context omitted.

https://en.wikipedia.org/wiki/Extended_evolutionary_synthesi... Essentially, variation comes from many other sources than random mutation on DNA. Random mutation itself is seen as mostly a bad source of variation, leading to destruction of the genome.

That's a very interesting hypothesis, thank you for the link. However, it doesn't seem to support what you're claiming about mutations, from my cursory reading. Can you provide a source for this claim that random mutation is a bad source of variation? http://extendedevolutionarysynthesis.com/about-the-ees/why-i... explicitly rejects the idea of a revolution regarding mutation and other genetic sources of variation: >…

There is John Sanford who claims random mutation is destroying our genome.

http://www.geneticentropy.org/

Re: Evolution Is the New Deep Learning

#195

Post by boomlinde »

photojosh wrote: post #157
flavio81 wrote: post #136

Earlier quoted context omitted.

>I'm amazed every time when the training is over and it actually works like intended. Everything which is big enough is more than the sum of it's parts. What about when it doesn't work as intended and fails ridiculously, even though it usually works perfectly well? http://www.labsix.org/physical-objects-that-fool-neural-nets...

Humans do the same thing; not working as intended some of the time. It's just that the failure modes for ML are different , and so we see them as ridiculous.

Well, if the result of the different failure modes means that a machine can't tell the difference between what obviously resembles a turtle and a rifle, or a cat and guacamole, then that's something that anyone who watched the video is better at. You can call it a different failure mode, but these are things no human would misclassify unless seriously ill, and being able to classify simple objects is important to our day to day lives.

Imagine some sort of Robocop deciding to neutralize someone for holding a turtle toy.

Re: Evolution Is the New Deep Learning

#196

Post by vanderZwan »

Houshalter wrote: post #161
vanderZwan wrote: post #153

Earlier quoted context omitted.

What you say is fairly hypothetical, since the way genetics works at a fundamental level prevents this possibility. You are also oversimplifying by not distinguishing local and global success. Locally, this gene will be successful. Globally, some mutations will be beneficial , and the children with those mutations will be more successful. The gene may become "dominant" (and then gain the beneficial mutations through…

It's not hypothetical. We haven't observed it in nature because any organism that evolved this would simply go extinct. But similar "evolving to extinction" phenomenons have been observed. Such as genes in mice that make the entire population male, or transposons in plants: http://lesswrong.com/lw/l5/evolving_to_extinction/ >it will not achieve a monopoly and not remove the existence of evolution. Yes it will. It's e…

To be clear on this: I agree with almost all of your reasoning (in this and the previous post), but this is not exactly the most convincing way to open your counter-argument:

> It's not hypothetical.

> We haven't observed it in nature

Also, the fact that in the long term such a mutation would cause a population to go extinct eventually is not really evidence that we should not find this in the wild, because it would still be effective in the short term. Look at that recent story about the mutated lobster taking over the waters of Germany for an example.

Anyway, the crucial disagreement lies here:

> It's easy to do simulations.

Yes, if a gene would evolve that brings the mutation rate to zero, this simulation works. But achieving such a thing sounds a lot like beating the laws of thermodynamics and stopping entropy from increasing. And sure, reducing entropy locally is possible by globally increasing it - in this case that would mean increasing the energy/resource budget spent on it, but I suspect that this in itself will be at a cost too high to give an advantage.

I basically don't believe (and I fully admit that this is a subjective point of view) that the "easy simulations" are noisy, large or complex enough to reflect the messy biological reality here. Essentially, there's too much approximation going on.

And btw, evolving to extinction is not the same thing: a gene that makes the entire population male is not in itself advantageous like reducing mutations is. It is more generic than the "mutation rate zero"-gene scenario, the latter is basically a specific hypothetical example of it.

Re: Evolution Is the New Deep Learning

#198

Post by jmmcd »

yters wrote: post #59

The only thing EAs have going for them is a biological metaphor, the magic of Darwinian evolution, fountain of endless novelty. But, modern science shows evolution does not really work in a Darwinian manner, so thus the metaphor ends.

> The only thing EAs have going for them

... is the ability to optimise a black-box objective.

Re: Evolution Is the New Deep Learning

#199

Post by jmmcd »

mav3r1ck wrote: post #15

Having studied this extensively back when they were called Genetic Algorithms, I would like to offer a few insights. 1) One of the biggest reasons they fell out of favor for more "mathematical" approaches was that no one could really explain why exactly they worked. It makes sense on the surface that "survival of the fittest" and doing something akin to multiple stochastic gradient descents would work, but no one has…

I think you've confused genetic algorithms with genetic programming. They're not the same thing.

> I don't know how Genetic Algorithms/programming could be made 'explainable' as to why they achieved an optimal solution other than hand-waving to how evolution works in nature.

This is quite confused. You're comparing different levels of the systems. In neural networks, we would like to know why a numerical model (which has been optimised by gradient descent) gives the outputs it gives. In GAs, (1) the objects being created usually aren't numerical models -- think instead of solutions to TSPs; (2) the reason the object is good is rather easy to see -- one just has to look at the objective function and verify that that the object has the desired properties; (3) we don't really care what other objects were considered during the search process.

Re: Evolution Is the New Deep Learning

#200

Post by jmmcd »

Cacti wrote: post #69
mav3r1ck wrote: post #15

Having studied this extensively back when they were called Genetic Algorithms, I would like to offer a few insights. 1) One of the biggest reasons they fell out of favor for more "mathematical" approaches was that no one could really explain why exactly they worked. It makes sense on the surface that "survival of the fittest" and doing something akin to multiple stochastic gradient descents would work, but no one has…

Yep. The original wave of genetic algorithms largely depended on some hand-wavy "building block" ideas that no one could really prove. It turned out that it was because proving them is impossible in the general sense, as we found out from the NFL theorems in the mid-to-late 90s, and it wasn't even clear the field had a scientific basis at all. So I was surprised to see them make a return about a decade later. Hopeful…

> make a return about a decade later.

Maybe that is how it looked from the outside. But the field continued as normal with no major problems, solving lots of industry problems with little hype.

Post reply on HN