binalpatel

12 ms·

91.
by binalpatel 10y ago
The interesting part of this is almost everyone was wrong in this (though when you're talking the realm of probability, it's hard to actually say they were wrong). In my statistics classes we often talk about bias and biased estim
92.
by binalpatel 10y ago
No formal sources unfortunately, it's similar to using a linear regression model trained on historical data for inference (with some cross-validation/penalization to make sure we're not just finding signals in the noise). We
93.
by binalpatel 10y ago
They have their place - if you're more interested in inference than in prediction they can be useful tools.
94.
by binalpatel 10y ago
I use AUC often, and always look at the confusion matrix. Metrics based off of it (recall, precision) are both useful to use as well. In the end the probability cutoffs you choose are more often than not based on business context and costs
95.
by binalpatel 10y ago
This is the imbalanced class problem that comes up often in the real world. You can't optimize for overall accuracy, otherwise you'd have classifiers that always just predict guilty (or not sick, or not fradulent, and so on) becau
96.
by binalpatel 10y ago
What if there's an explosion of risk taking and creativity? What if anyone who ever wanted to start a business, goes for it, instead of taking a stable job because they're free to do so? We won't know what'll happen till
97.
Planet Earth II Trailer(youtube.com)
3 points by binalpatel 10y ago|0 comments
98.
by binalpatel 10y ago
Situations like this really let you know the true merit of your leadership. I had a similar situation - where alerts based on bad data went out to customers, and hundreds of customer support tickets started flooding in. The VP, Director, an
99.
by binalpatel 10y ago
I guess my frame of context is, it's their money, and they can spend it however they may, but at least with this there is a chance of benefiting society as a whole. While your example seems fitting at first glance, I don't really
100.
by binalpatel 10y ago
I don't understand the vitriol. They may, or may not, reach their goal, but it's certainly worth trying. Even if they don't cure all diseases - if they advance our understanding of some of them, or help cure some, then humani
101.
George Soros Will Invest $500M to Help Migrants and Refugees(reuters.com)
1 points by binalpatel 10y ago|0 comments
102.
by binalpatel 10y ago
Largely the first mover advantage, I think. I'm sure once someone does it it'll be a (or it already is) a gold-rush, but the first to get to the finish line will realize the most gains initially. I'm also not sure how relevan
103.
by binalpatel 10y ago
You clearly have a viewpoint you're entrenched in - and you're pigeon holing this into that. Uber is a bad investment to you - but a good investment to others. If in the end they successfully become one of the first firms to have
104.
U.S. Household Income Grew 5.2% in 2015, Breaking Pattern of Stagnation(nytimes.com)
5 points by binalpatel 10y ago|0 comments
105.
by binalpatel 10y ago
I love this, it seems like we've been pushing the frontier of the very large and the very small in the last few years (space scale and molecular scale), and I can't wait to see what come out of it.
106.
by binalpatel 10y ago
The fact that the original, unmodified article referred to data engineers as "janitors" pretty much says it all. It's very analogous to front-office and back-office work in Investment Banking. "Data Scientist" are t
107.
by binalpatel 10y ago
I used the app daily for a full year at one point, and really loved it. Rewatching it - 6 minutes for an entire year, was amazing. You forget so much as time goes on, having those seconds spark memory upon memory is more than worth it.
108.
by binalpatel 10y ago
They're sending up a CubeSat which looks to be pretty small, from Googling it looks like the cost to launch could be $40-60k USD. I'm not familiar with testing things like this, but I could very well see that being cheaper than hi
109.
EmDrive Will Be Tested in Space(popularmechanics.com)
4 points by binalpatel 10y ago|2 comments
110.
by binalpatel 10y ago
Found this pretty informative answer to one of the questions I had (mainly, whether Zika had always been this bad, or had mutated in some way to make it worse). http://biology.stackexchange.com/questions/42891/did-
111.
by binalpatel 10y ago
Start with a higher level interface and work backwards? Possibly something like Hive/PIG/MRJob, get familiar with them to wrap your head around MapReduce. Past the scope of your question - but I'd also recommend learning Spar
112.
by binalpatel 10y ago
Humans do exactly that though - an artist isn't born in a vacuum.
113.
by binalpatel 10y ago
We have proof that consciousness and insight are possible in nature - mainly us. We replicate it with each generation, and watch it grow as children age.
114.
by binalpatel 10y ago
One of the reasons I chose to settle down in Raleigh, NC. Not a large city, but a great tech scene that still pays well, and very reasonable cost of living. Not as much to do as San Francisco or NYC - but I certainly think it's worth i
115.
by binalpatel 10y ago
Initial fee of 2-3% of the balance to transfer, then 0% APR for a 1 to 1.5 years normally.
116.
by binalpatel 10y ago
I'm probably an odd one out, but I love using credit cards. Got my first student card as a freshmen in college, and at this point I have about 10 cards, each with a very high limit (most of them with nothing on them). I treat them like
117.
by binalpatel 10y ago
The Rational Optimist. It helped me realize (along with other things), that no, the world isn't getting worse. Things are much better than they've ever been for most people, and they're only getting better, faster.
118.
by binalpatel 10y ago
Jvion | Atlanta, GA | Relocation | ONSITE At Jvion we use machine learning and data science to help improve patient outcomes. We have our solution live in several hospital systems, and help hospitals identify patients who'll have bad o
119.
by binalpatel 11y ago
Jvion | Atlanta, GA | Relocation | ONSITE At Jvion we use machine learning and data science to help improve patient outcomes. We have our solution live in several hospital systems, and help hospitals identify patients who'll have bad o
120.
by binalpatel 11y ago
Glad you came to the conclusion that I always seem to come to, it's all about the features. Engineer great features and you'll get great results, bad features and you'll have bad results.

More ›