Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

This is the basis of "artificial creativity". Train these NNs to recognize faces (or anything else, maybe especially everything else), and then run random noise through them. See what they come up with.

You could have an algorithm coming up with some pretty decent (and original) cartoon faces. Or with art, abstract or not.

This is what your brain does.



Now we just need a 90s Drug PSA spoof,

"This is your brain. This is your brain on neural networks. A cloud is not a face. Neural nets, not even once."


Here is an implementation of this approach from a few years ago:

http://iobound.com/pareidoloop/

I experimented with doing something similar, trying to generate faces from random line drawings, but I didn't get very far.


That doesn't sound right. You suggest if you give the computer something that looks like a face, then add a filter, it has somehow created an idea on its own. Yet you are still telling the computer what to do, it has done nothing creative on its own. I'd venture that real "creativity" is the reverse ... the ability to create something that is recognizable as a face.


No. I suggest giving the thing noise. Give it lots of noise.

Let it find its own face. Let it find several, and let it score the ones it thinks are the best.

Then maybe toss those at another NN that's been trained on cartoon faces. Let it modify this noise-face so that it's more cartoonish.

And just keep iterating through it like that.

> it has somehow created an idea on its own.

Can we please grow up. Creative people aren't "creating" like some judeo-christian deity ex nihilo.

They're exploring a finite ideaspace and keeping the best ideas they stumble upon.


There's an entire branch of computer science devoted to studying creativity and computation. Google Prof Geraint Wiggins for some examples.

His ideas about this are a bit more thoughtful and insightful than 'exploring a finite ideaspace.'


If he wants to make a career out of something so simple, I'm not going to stop him. Why anyone would listen to him, that's another thing entirely. Perhaps you like the sound of bullshit.


Random input filtered by face detection does create new faces. It's just a different method of creating faces than one that derives a face from the parameters of the face filter (which is more like how a human would do it), but either way, you get new output that is identifiable and identified as a face.

Yes, the computer is biased by your input of the face detection filter, but how is that any different than a human's cultural biases? The only real differences are purely internal details.


Read up on restricted boltzmann machines and their ability to dream.


For me, creativity often happens by getting a random idea and recognizing it as part of a whole which is yet to be created. In the linked case it could be something like identifying a face in the clouds and then extrapolating that to create an image of a whole cloud person.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: