cardigan

10 ms·

31.
Expert prediction of social science experiments(slatestarcodex.com)
1 points by cardigan 10y ago|0 comments
32.
by cardigan 10y ago
Doomsday prepper websites
33.
Uber Elevate: Fast-Forwarding to a Future of On-Demand Urban Air Transportation(uber.com)
4 points by cardigan 10y ago|0 comments
34.
I Went Undercover with a Border Militia(m.motherjones.com)
4 points by cardigan 10y ago|0 comments
35.
by cardigan 10y ago
The channel does not need to be secure, only authenticated. So e.g, you could send them a Slack message then have them call you to confirm you just sent that link and not someone who hacked your Slack account. It's fine to have someone
36.
by cardigan 10y ago
Sorry that's not the truth, I read about a book a week. This week I've read 3 and started another 2. The secret is to read for enjoyment, a lot, and eventually you get fast. If it feels like a chore, you're not reading the ri
37.
by cardigan 10y ago
Sorry I don't understand your question. Obviously very few people living in Kashmir support those groups, see polls.
38.
by cardigan 10y ago
I did say most people living in Kashmir. Last time I checked, Kashmir wasn't Jammu or Ladakh.
39.
by cardigan 10y ago
Starts off with an interesting idea (programming is perfectly specified versus being underspecified but more efficient due to shared context) but then devolves significantly: neural networks are not the be all end all of smart computers. I
40.
by cardigan 10y ago
It's promoting terrorism if you're politically aligned with India. If you're instead politically aligned with most people living in Kashmir, it's not promoting terrorism, you see it instead as showing solidarity with the
41.
by cardigan 10y ago
Less than a day for Ready Player One ;( Not focus for me as much as how much fun I'm having
42.
by cardigan 10y ago
Because income inequality is more important to this guy than providing for people's basic needs lol
43.
Reverse engineering slot machine firmware for $1M in winnings(ethicalhacker.net)
3 points by cardigan 10y ago|0 comments
44.
by cardigan 10y ago
Largely automated delivery infrastructure - pipes between places, or entry hatches in roofs to inside offices/homes (drone delivery) and hatches within buildings to some storage area within the building are required the way other infra
45.
by cardigan 10y ago
So will Webpass be faster now?
46.
Gigster Fund – equity comp hack for 1099 contractors(medium.com)
2 points by cardigan 10y ago|0 comments
47.
De Beers Diamond Report: "fact that they are seen as a lasting source of value"(insightreport.debeersgroup.com)
1 points by cardigan 10y ago|0 comments
48.
by cardigan 11y ago
Reading this while visiting family in Srinagar :-) Lots of brilliant young people who speak English but end up working dead end jobs because of the political situation's impact on the economy. And they want to live with their families,
49.
by cardigan 11y ago
cool to see this open source - readme talks about versioning; is that like deployment versioning for kube objects? how is that different than what's coming out in 1.2?
50.
In Praise of Idleness, by Bertrand Russell(harpers.org)
55 points by cardigan 11y ago|25 comments
51.
by cardigan 11y ago
Facebook M? Magic?
52.
by cardigan 11y ago
Use the same code but with a list of company names?
53.
by cardigan 11y ago
The network learns something about the relationship between characters, and by extension phonemes, in the training set of nice sounding words. Obviously it could not learn the image the word invokes, given the size of the training set and s
54.
by cardigan 11y ago
The first kind of scoring function was language independent (e.g, penalize words showing up from the training set, penalize < 3 character words, penalize duplicate words, penalize common words), and the second scoring function needed a l
55.
by cardigan 11y ago
Aww :-) you're very welcome!
56.
by cardigan 11y ago
This is for brandable names I guess? That sounds like a good approximation: although it's more judging if the word is spelled like it sounds versus if it has just one obvious spelling. Similar ideas: check social media account availabi
57.
by cardigan 11y ago
Larger corpus, deeper model, better scoring function on models, more models
58.
by cardigan 11y ago
It's using either an RNN/GRU/LSTM: realized this after posting; one of the parameters for the models being trained is the kind of model, and it picked randomly between an RNN, GRU, and LSTM. Not sure what the five models pick
59.
by cardigan 11y ago
Yes totally can be. Nothing English language specific about this. Scoring the models is a bit of a pain without a large corpus of English, but the training data was only 800 words.
60.
by cardigan 11y ago
Ooh front page; I guess this calls for a bit of an explanation! First off - used this code for training the models: http://karpathy.github.io/2015/05/21/rnn-effectiveness/ Very very easy to setup and tra

More ›