slashcom

8 ms·

31.
by slashcom 9y ago
They can’t say that though, because recording a video involves accessing the microphone. Otherwise Facebook Live and Stories couldn’t exist.
32.
by slashcom 9y ago
It's like, fine. If you're an engineer curious about the basic notions of deep learning, have at it. It's not very technical; very big picture. If you're well experienced with deep learning already, maybe only the last 2
33.
by slashcom 9y ago
It's worth noting this post caused a very intense twitter debate among the NLP and DL communities, especially after Yann Lecun replied to Yoav's comments. https://www.facebook.com/yann.lecun/posts/1015449
34.
by slashcom 9y ago
The *ACL conferences (major NLP venues) decided to allow presentations to be in Mandarin now; the rationale being that a lot of presentations were in such poor English already, might as well let half the audience be engaged. It will take at
35.
by slashcom 9y ago
I worry that they might sell it as a reason backdoors are necessary: if only we had backdoors, we could've saved those patients! The flaw of this logic would be lost on most lawmakers.
36.
by slashcom 9y ago
Real publishers make a good chunk of their profits from selling subscriptions of their journals to libraries. These allow anyone at a university to read the journal, but the fee can be many thousands per year (per university). These are leg
37.
by slashcom 10y ago
df is a dataframe, which is like a spreadsheet. This line takes the logarithm of the column named 'y' and updates it in place.
38.
by slashcom 10y ago
It wasn't due to demand though. It was due to price being calculated by sensors that detect lane speed. The lane was ridiculously slowed by two snow trucks, so it computed a high price. That said, it's probably reasonable snow tru
39.
by slashcom 10y ago
Your mediocre graphics card may or may not have CUDA support. If it doesn't, or it only supports an old version, then it's the same as not having a graphics card at all. And of course, Nvidia only. If it supported, then the major
40.
by slashcom 10y ago
Probably not for a lot of the state, but the Charlotte metro area contains a significant number of people who live across the border in Rock Hill for property tax reasons. These people make that commute almost daily. (It's actually eno
41.
by slashcom 10y ago
Interestingly, the difference in SC sales tax (6%) and NC sales tax (4.75%, but county/city taxes make it ~6.75%), effectively come out to nearly $20 on a $2000 laptop.
42.
by slashcom 10y ago
But if you have it tied into an investment vehicle, then you're taking some risk and putting that wealth to use.
43.
by slashcom 10y ago
http://lbzip2.org/ also exists for bzip2. It works really well, especially since bz2 files are split into discreet blocks which can be un/compressed independently.
44.
by slashcom 10y ago
Forgive me, but I'm less impressed by the paper. As far as I can tell, they've only really shown that (1) language is recursive, which we know already; (2) markov models cannot capture recursive languages, which we've known;
45.
by slashcom 10y ago
According to https://en.wikipedia.org/wiki/Demography_of_the_United_State... 27.1% of the US is under 21 and 14.5% is over 65. I know 16 should be working but a good chunk should be getting education. That leaves only
46.
by slashcom 10y ago
Will be interesting to see if they pull out like they've done in Austin. It's definitely been less convenient to go to a bar since they left.
47.
by slashcom 10y ago
Yeah, it's written like if this law doesn't pass, China will be have full access to the NSA.
48.
by slashcom 10y ago
Pretty good; 8gb memory and impressive transfer speeds. The 980 Ti is pretty common in the Deep Learning community. Only the Titan X is more popular, and that thing is far outside any average person's budget. I'm definitely lookin
49.
by slashcom 10y ago
http://www.nature.com/nature/journal/v532/n7600/full/nature1...
50.
by slashcom 11y ago
Most interesting to me was that the gender gap was so much more pronounced in SV than outside non-SV male median: 97000 non-SV female median: 90000 (92% of male) SV male median: 137120 SV female median: 99187 (72% of male!) Unless there&#x2
51.
by slashcom 11y ago
Just left a DARPA grant for an AI project. They've been one of the biggest AI funders for some time.
52.
by slashcom 11y ago
My guess as to what FB will do: treat them all as a "like" with respect to engagement statistics (i.e. do we show this post to more people?) And simultaneously, use all the emotions to train a ridiculously good sentiment analysis
53.
by slashcom 11y ago
I can't speak as to how to individual particular programs work, but at least as of a year ago, most of the good AIs used something called Monte Carlo Upper Confidence Trees. They're, unfortunately, not based on intuition. Just sta
54.
by slashcom 11y ago
So this is my research area. Distributional vectors are a proxy to word meaning. Words with similar vectors have similar meaning or are semantically related in some way. But usually, you just measure similarity by a single number: the cosin
55.
by slashcom 11y ago
http://www.datascienceweekly.org/ The editor picks are usually extremely good, and if I didn't read them already, I know that I should.
56.
by slashcom 12y ago
Bleach it?
57.
by slashcom 12y ago
Karpathy constantly shows the gap between "Anyone could've done that" and "Yeah, but he _did_."
58.
by slashcom 12y ago
The original version of the paper was what Yoav was criticizing. It's worth noting that the authors made nontrivial changes to address most of Yoav's comments (and as a result, ended up with a much higher quality paper). I special
59.
by slashcom 12y ago
The source of the data also contains many interesting comparisons and conclusions, particularly with respect to time. http://www.nsf.gov/statistics/seind14/index.cfm/chapter-5/c5...
60.
by slashcom 12y ago
Same; though it looks like it might be because my account is a Google Apps account.

More ›