zan2434

This is unbelievably sad. What a tragic existence.

  Beneath, his skin is raw, revealing red ulcers caused by his disease. “Itai,” he says. It hurts. Then he smiles. “Gambatte,” he says – I will endure.

More than a little disingenuous. While they advertise that the app supports a number of services, when you actually download the app and try it only Slack and Skype are available - the rest are "coming later in February"

This is amazing! This free "cardboard" digital camera from IKEA is particularly cool: http://old-digitalcameras.com/MorePicts/MP656.htm

I don't understand. How does it respond to your queries? For weather it seems to animate little rain droplets on the screen, but in the example: "what's an epsilon?" they never show it actually answering.

doesn't PCB stand for "Printed Circuit Board"? So it's still mostly printed.. The rotor is 3d printed, and the stator is manufactured via PCB techniques. I understand what you mean regarding it not being totally printed though.

Related article on similar hierarchical / compositional policies learned by maximum entropy optimization: http://bair.berkeley.edu/blog/2017/10/06/soft-q-learning/

Although it is quite egregious here - this is not a problem inherent to colorization but rather to generative models in general.

Using something akin to a variational autoencoder would solve this problem, because it learns a distributional approximation rather than a single point estimate of the color, and then the random noise vector input allows one to sample from this output distribution. Similarly, Mixture Density Networks allow you to model a distribution and then sample from it.

Data scientist at Opendoor here. That is a very accurate analogy.

This is awesome! Can you describe how you implemented the WebGL ops a bit more? Did you have to write your own convolution kernel with GLSL for example?

Based on the rest of the comments here deriding the growth over revenue strategy I think it's very important to bring up that risk is proportional to reward, and by definition any business that can be cash flow positive early on is unlikely to be very risky - and thereby not really what VCs are in this business for.

Garry was endlessly supportive and encouraging through all the tough times at YC. He'll be missed!!

http://deepdreams.zainshah.net spun up a simple web server so you can try your own! Please be gentle :)

or just one quadcopter, with one camera, that moves around the object

Hmm I always default to not "do what you love" but "love what you do" in that most work has some interesting component, and what's really important to develop is the ability to isolate that aspect and motivate yourself with it. That's something I think everyone can use, regardless of their position in life

was about to post this exactly. Looks like it really is closer than it appears!

This was really cool! I realize there are many DoF in this so pure optimization likely wouldn't solve it, but I'd like to see an optimization on the form of the keyboard itself in addition to just the ordering of the keys (like minuum?)

TL;DR: In the context of our existence, only the brevity of our lives strikes us as absurd.

I was really hoping this was going to be an AR hack for seeing WiFi, but this is cool too!

>Perhaps a neural network can tell what a cat looks like, but can it tell the difference between a dead cat and a sleeping one, by watching it breathe?

Yes, yes it can: http://arxiv.org/abs/1411.4389 (Long-term Recurrent Convolutional Networks for Visual Recognition and Description) It's certainly surprising how much of perception can be reduced to classification.

I wonder if this concept (corneal reflection photography) can be used for an eye-tracking interface for computers. Can you match the segment of the image of the corneal reflection that is above the pupil with what is on the screen to figure out where/at what the user is looking?

Wait he just claimed the universe is computable because it produces regular data we can encode. Is that necessarily the case? Can't you claim that our senses just produce regular data that we can encode?

Despite the article, I think the title's premise is very interesting. The resurgence in popularity of dystopian sci-fi really seems to me reminiscent of Peter Thiel's "indeterminate pessimism" - the attitude that the future is inevitably bad. This is a terribly unproductive attitude to have, while the opposite attitude of "determinate optimism" characterizes us during times of great progress.

I wonder if being open minded is enough. The "winds of change" aren't always very strong, and accordingly new opportunities aren't (at least initially) better alternatives to the status quo, they just happen to have incredible future potential. To fully take advantage of this, the expert would have to move backwards, quite possibly endangering himself, while a new entrant to the field has no risk to assume by doing so. To use the hill-climbing metaphor described by other commenters, a new climber can begin climbing the new potentially tall mountain immediately, while the "expert" will first have to descend his peak.

I don't see how "why" can ever definitely be resolved. It's far too nuanced a question, but higher resolution data capture and analysis helps you understand it more easily, which seems to be the direction analytics is going.

This is so clever! Turning mechanical control problems into informational control problems is critical to the ubiquity of micro air vehicles.

That was beautiful. I'd love to read more about how exactly the visuals were generated. This is a lot more than spheres rendered w/ NASA photo textures.

This is brilliant. Lyft is doing everything right as of late.

just added function previews to make playing easier!

source here: https://github.com/igul222/Axis