reiinakano

10 ms·

1.
by reiinakano 7y ago
but they can all invert a binary tree on a whiteboard, so all good
2.
by reiinakano 7y ago
No, it doesn't. Shared weight means all weights are the same value, as I originally mentioned.
3.
by reiinakano 7y ago
Neither. The architecture performs well independently of weights BUT all weights must be the same e.g. Same network works well when all weights are 5.0 or when all weights are -3.0
4.
by reiinakano 7y ago
> I'm much less confident about everything This doesn't seem like a bad thing. Maybe people just realize as they grow older how complex things really are, and eventually learn not to fall for the dunning Kruger effect on everyt
5.
by reiinakano 7y ago
I did a search for Flamingo and there were no flamingos...
6.
Intrinsic Neural Style Transfer: Style Transfer Without a Style Image(twitter.com)
4 points by reiinakano 7y ago|0 comments
7.
Intrinsic Neural Style Transfer: Style transfer without a style image(twitter.com)
2 points by reiinakano 7y ago|0 comments
8.
Intrinsic Neural Style Transfer(twitter.com)
5 points by reiinakano 7y ago|0 comments
9.
Intrinsic Neural Style Transfer(twitter.com)
6 points by reiinakano 7y ago|0 comments
10.
Ask HN: Machine learning contractors, how much do you charge hourly?
94 points by reiinakano 8y ago|40 comments
11.
Teaching neural networks to use a paintbrush(reiinakano.github.io)
2 points by reiinakano 8y ago|0 comments
12.
Teaching neural networks to paint inside their own dreams(reiinakano.github.io)
2 points by reiinakano 8y ago|0 comments
13.
Teaching agents to paint inside their own dreams(reiinakano.github.io)
1 points by reiinakano 8y ago|0 comments
14.
Porting arbitrary neural style transfer to JavaScript(magenta.tensorflow.org)
2 points by reiinakano 8y ago|0 comments
15.
by reiinakano 8y ago
Cause you can do things like this https://magenta.tensorflow.org/blog/2018/12/20/style-transfe...
16.
Porting arbitrary style transfer to the browser(magenta.tensorflow.org)
3 points by reiinakano 8y ago|0 comments
17.
Show HN: Paint yourself in the style of classic art, right in your browser(reiinakano.github.io)
6 points by reiinakano 8y ago|1 comments
18.
by reiinakano 8y ago
Yes, a lot of other apps and websites have been doing this for years. But you have to upload your images to their server. The point of this is that it does all calculations locally in the browser, at the expense of some quality.
19.
by reiinakano 8y ago
That's a good observation. Theoretically, 0 strength should give you the content image. What I think is happening is that the algorithm is not trained to generate style vectors for photorealistic images, and so the mapping it learns fr
20.
by reiinakano 8y ago
Hi HN! A year ago, I shared a demo where you could paint a photograph in the style of one of five paintings, via neural networks, right in the browser. Since then, a few papers have come out attempting to create a single model that could be
21.
Show HN: Play with arbitrary neural style transfer directly in your browser(reiinakano.github.io)
3 points by reiinakano 8y ago|3 comments
22.
Show HN: Paint any photo in the style of any painting, directly in your browser(github.com)
46 points by reiinakano 8y ago|8 comments
23.
by reiinakano 9y ago
"Theranos, diagnose diseases with AI and Blockchain support"
24.
Harassment culture in academia(twitter.com)
1 points by reiinakano 9y ago|0 comments
25.
by reiinakano 9y ago
> I don't think the author has any insight at all on what a general AI will be. To be fair, nobody does. But you're right, the author shouldn't be making these statements with such certainty.
26.
by reiinakano 9y ago
Sure, they do their best to give an explanation after discovering it works, but nobody ever publishes the hundred other models they tried that don't work out.
27.
by reiinakano 9y ago
Looks about right. The default setting is as basic as you can get. No transpose convolutions, batch norm, and other tricks for better GANs.
28.
by reiinakano 9y ago
Unlike regular classification, it's hard to quantify the performance of a GAN by just looking at the loss curves. You really have to make a subjective evaluation by looking at the generated images. By playing around with it, the result
29.
by reiinakano 9y ago
Default parameters on MNIST should give good results in 5 minutes give or take - depending on your hardware. Can you show what you got?
30.
Show HN: GAN Playground – Play with Generative Adversarial Neural Nets In-Browser(reiinakano.github.io)
97 points by reiinakano 9y ago|14 comments

More ›