leblancfg

30 ms·

271.
You Can Finally Train Neural Networks on AMD and Intel GPUs (Torch)(arxiv.org)
3 points by leblancfg 10y ago|0 comments
272.
by leblancfg 10y ago
I really know nothing about this line of research AT ALL. It's all quite interesting, thanks for linking to it. Now, the general rule of thumb for drugs is there's no such thing as a free lunch , what are the possible negative ou
273.
by leblancfg 10y ago
The history of concrete is quite interesting. http://www.romanconcrete.com/docs/spillway/spillway.htm http://io9.gizmodo.com/how-the-ancient-romans-made-better-co...
274.
How to Conquer Tensorphobia (2014)(jeremykun.com)
1 points by leblancfg 10y ago|0 comments
275.
by leblancfg 10y ago
Yup.
276.
by leblancfg 10y ago
His name was Satoshi Natamoto.
277.
by leblancfg 10y ago
Well, Wolfram also spent time with Minsky, so he'd have stories to recount, but not with Ramanujan. That alone would explain the numbers. I find the Ramanujan story actually reads more like a Mathematica advert, whereas the other more
278.
by leblancfg 11y ago
I agree that it sounds like lobbying. That being said, I think illusionists are definitely practitioners of an art form in its own right, so, good on them. > Whereas magic enables people to experience the impossible Now you're just
279.
by leblancfg 11y ago
A large part of theoretical physics is getting to grips with the terminology -- not that you should imply that I'm in any way an expert. I think I'd go with the wording "it seems more like a change in the way we understand
280.
by leblancfg 11y ago
Mr. Vorhies seems to have forgotten to take into account the adoption curve in his predictions -- not that I think his predictions are wrong, but I think 2025 is waay too early for that change to take effect to anyone other than the 1%, b
281.
by leblancfg 11y ago
First off, credit goes where credit is due, that is great research from the team at Cambridge. If proven further, that evidence is sure to change every evolution course taught in the future. For me though, that headline just killed the st
282.
by leblancfg 11y ago
Interesting idea, but nothing seems to work here, both on Firefox and IE 11. All I can see is the left "control pane".
283.
by leblancfg 11y ago
I'm a neophyte here, but I don't understand what that means. Are you asking past YC founders to release a short statement regarding "starting-up understanding"? And what are cliffs?
284.
by leblancfg 11y ago
This. And it's "Tahr", goddamnit. XD
285.
by leblancfg 11y ago
I thought this also, but "[...] to keep the beam tightly focused on one probe at a time would require [...]" in the original article points in the opposite direction.
286.
by leblancfg 11y ago
Well, that sure is a good point. Upon further reading, looks like you can achieve >99.9998% reflectance for a 10% wavelength band (in this case, in the infrared). [ http://www.crystallinemirrors.com/wp-content/uploads
287.
by leblancfg 11y ago
I love these napkin calculations. The relativistic effect shifts the frequency by 22.47% for 0.2c. Given a 445nm laser (the powerful blue ones), that's 545nm when redshifted (green). It's a bit of a jump, but that wavelength is st
288.
by leblancfg 11y ago
I agree. And certainly don't think MIT, nor Harvard, are shit. You get what you put in. Mostly, I was educated in the Québec system, and had no idea before I was 16 that if I got good grades I could get into interesting universities
289.
by leblancfg 11y ago
Reading this made me downright sad that I didn't, and probably never will, attend MIT. Oh, the interesting people! The interesting projects! How I wish I had worked harder.
290.
by leblancfg 11y ago
I think you taking the high road and aim for the "gain this startup as a Chinese subsidiary" route is the win-win situation here. Of course, I can only read so much from your description, and maybe some things don't apply in
291.
by leblancfg 11y ago
...but this is a country. In the immediate, the only thing that can happen, if at all, is for some people to lose their jobs. I think he is hoping that if the leak is well covered enough by the media, it will be adding oil to the fire of pu
292.
by leblancfg 11y ago
>[W]e should be able to assume that they obey the physical laws of the universe I don't think anyone is assuming otherwise... all I see is statistics and sampling being argued. You'll have to endure my wall of text, though, sor
293.
by leblancfg 11y ago
My only gripe is that he's only used vertebrates in his dataset (based on http://journals.plos.org/plosone/article?id=10.1371/journal.... ). It's "only human" to include close members on the phy
294.
by leblancfg 11y ago
>> Companies aren't libraries. They aren't paying you to sit and read books. That stroke you've brushed is too wide. Smart employers will have some of the money they're paying an employee going towards learning...
295.
by leblancfg 11y ago
From a remote sensing standpoint, this sounds a bit ridiculous. Methods on how to map roads and streets to GIS vectors from (public domain) hyperspectral satellite images have been around for at least 20 years. Technology-focused job creati
296.
by leblancfg 11y ago
Could be a future feature for airbnb.