unixpickle

7 ms·

31.
by unixpickle 7y ago
Good idea, I'll add a reference to SDFs. Didn't think about the normal-via-autodiff aspect, which would be quite useful. It seems to me that SDFs are very hard to program from scratch, compared to simple boolean functions. Obvious
32.
by unixpickle 7y ago
This is a web demo for creating 3D models in JavaScript. It uses a variant of ray casting that can render any function mapping `(x, y, z) -> bool` in real-time. It's easy to define very interesting 3D models as boolean functions. In
33.
Show HN: Render Arbitrary 3D Boolean Functions(aqnichol.com)
8 points by unixpickle 7y ago|6 comments
34.
by unixpickle 7y ago
I've used vanilla Ubuntu on my Dell XPS for the past year and it has served me well.
35.
by unixpickle 7y ago
Maybe we are just preparing for the point when climate change makes going outside difficult and largely unsafe...
36.
by unixpickle 7y ago
Of course, google in particular is behind a very complex distributed network. Distributed DBs are mentioned, but it would be cool to know more about how web requests are distributed and routed in this system.
37.
by unixpickle 8y ago
Not the first time someone left a MongoDB database exposed to the wild, and it won't be the last. It's an easy thing to do, especially since MongoDB is so popular for small single-server projects. A few years ago, I discovered the
38.
by unixpickle 8y ago
> Also, ImageNet’s website was down on the day that we decided to perform this research. We therefore paid an unemployed person to download 100 examples of paint-splatter artwork by searching Pinterest using the query “paint splatter wal
39.
by unixpickle 8y ago
Reproduced the bug in Golang. Pretty simple stuff. https://github.com/unixpickle/cve-2018-4407
40.
by unixpickle 8y ago
Even _MNIST_ with a decent CNN comparison would be better than what they have here.
41.
by unixpickle 8y ago
I don't understand how papers like this end up on the homepage of HN. The abstract makes very broad claims, but the paper itself only runs toy experiments on tiny datasets with tiny neural networks. No good ML practitioner believes tha
42.
by unixpickle 9y ago
In terms of "backpropagating through thousands of frames", it's not as expensive as you might think. I've used TRPO to train RNNs on games like Atari pong with thousands of frames per episode. This can be done via an alg
43.
by unixpickle 9y ago
(Author here). Hi Greg! I am excited to hear about the new Universe environments. I want as many RL environments as possible for my upcoming project, so I will probably draw from Universe and ALE as well as µniverse. I took a lot of inspira
44.
by unixpickle 9y ago
(Author here). I hadn't realized that the engineers were laid off. Where did you find this out?
45.
by unixpickle 9y ago
I'm pretty sure that was the goal of OpenAI Gym. Gym tries to provide a generic interface for RL environments, and imho it does a nice job. I am working on Python bindings for µniverse now, which should allow µniverse to integrate with
46.
by unixpickle 9y ago
I'm not sure I see why tanh couldn't be used to the same effect. If you use 1.6*tanh(x) as your activation function, it pushes small variances higher and high variances lower and gets you to a variance of ~1 after many layers. Obv
47.
by unixpickle 10y ago
The stuff in Appendix C worries me a little. Dividing the numerator and denominator indeed does not change the output at timestep t. However, if you use the new, smaller numerator and denominator at later timesteps, later outputs will be
48.
by unixpickle 10y ago
Ah, now it's much more clear. By the way, have you tested the model without parameterizing z the way you did? Different z's might be worth further exploration.
49.
by unixpickle 10y ago
I'm unclear about why z needs to be parameterized in terms of two functions, one unbounded, u(x), and a one bounded, tanh(g(x,h)). Shouldn't the squashing function f() prevent explosion anyway? And how would this parameterization
50.
by unixpickle 14y ago
Right now I'm working on a command-line utility for ripping albums off Youtube. It uses some music database APIs to lookup tracks for a specified album, and then finds individual youtube videos to rip for each track. Once a video is downloa
51.
A Character Encoding Glitch In Messages.app(jitsik.com)
3 points by unixpickle 14y ago|0 comments
52.
One blogger's expectations of WWDC(thoughtsfromanipad.com)
2 points by unixpickle 14y ago|0 comments
53.
Minimalism & simplicity - the right choice for you?(whatitellmyself.tumblr.com)
2 points by unixpickle 14y ago|0 comments
54.
Higher learning(thoughtsfromanipad.com)
3 points by unixpickle 15y ago|0 comments
55.
by unixpickle 15y ago
You could also use objdump -d to decompile the binary.