News for hackers. Stuff that matters.
Re:Machine Learning by Houshalter on Wednesday May 17, 2017 @05:56AM (#14356962)
Reminds me of this horrifying stack exchange post: https://stats.stackexchange.com/questions/185507/what-happen...
Re:Machine Learning by sevensor on Wednesday May 17, 2017 @09:22AM (#14358117)
One of my colleagues (an engineer) suggested something like this when I worked at a factory. My jaw dropped and I just stared at him. I had nothing to say. I guess he's a manager now.
Re:Machine Learning by cleetus on Wednesday May 17, 2017 @09:35AM (#14358229)
I can't believe that's real!
Re:Machine Learning by SBCRec on Wednesday May 17, 2017 @06:46AM (#14357191)
wow, that is actually horrifying.
Re:Machine Learning by jmkni on Wednesday May 17, 2017 @07:09AM (#14357310)
ELI5?
Re:Machine Learning by TeMPOraL on Wednesday May 17, 2017 @07:23AM (#14357378)
The manager of the person who asked the question thinks that if you take data in form of pairs (X, Y), split them up, sort independently and combine again, you'll get better results. In fact, such operation obviously destroys the relationship X had to Y, so the result is meaningless.
Re:Machine Learning by jmkni on Wednesday May 17, 2017 @07:27AM (#14357399)
Ha, gotcha! Maybe if you do it enough times...in the cloud?
Re:Machine Learning by ardivekar on Wednesday May 17, 2017 @07:48AM (#14357507)
No no, results stored on The Blockchain.
Re:Machine Learning by pavel_lishin on Wednesday May 17, 2017 @09:38AM (#14358254)
A chatbot can read them back to us.
Re:Machine Learning by sidlls on Wednesday May 17, 2017 @11:16AM (#14359250)
Only if it's implemented in Rust. Any other language wouldn't be safe.
Re:Machine Learning by yellowapple on Wednesday May 17, 2017 @02:53PM (#14361634)
But Rust is not Web Scale™. Go.js is Web Scale™.
Re:Machine Learning by TeMPOraL on Wednesday May 17, 2017 @07:41AM (#14357478)
You'd be better off just going with whatever "result" you want regardless of the data, and save on the AWS costs.
Re:Machine Learning by Buttons840 on Wednesday May 17, 2017 @10:27AM (#14358722)
Doesn't it also assume you'll have access to the correct answer in production as well? Is X the observation and Y the correct answer, as often indicated by that notation?
Re:Machine Learning by nilkn on Wednesday May 17, 2017 @12:14PM (#14359894)
Suppose your dataset consists of the following points.

(1,5)

(2,3)

(3,10)

(4,4)

Clearly there's really not a linear relationship between the first coordinate and the second coordinate. In other words, if you plot these four points and try to approximate them with a single line in the plane, you just can't do a very good job.

The person's manager suggested sorting the second coordinates in this list while keeping the first coordinates fixed, resulting in the following:

(1,3)

(2,4)

(3,5)

(4,10)

These points are still not collinear, but they certainly can be better approximated by a line than before. The problem is that this is simply a completely different set of points, so a linear approximation here implies nothing about the original dataset.

Re:Machine Learning by chestervonwinch on Wednesday May 17, 2017 @08:42AM (#14357825)
Imagine you were trying to measure whether or not increasing X proportionally increases Y. Now imagine you first sort your data points, Xi and Yi, independently -- it will now misleadingly appear like Y increases when X increases since the variables are both sorted. You've scrambled your data (Xi, Yi) to be (Xk, Yj). So you're essentially working with a totally different dataset now.
Re:Machine Learning by jmkni on Wednesday May 17, 2017 @11:40AM (#14359508)
It's actually quite a simple question, just asked using a lot of overly complex language which probably confused the questioner and his boss.

It's no different to having a list of key/value pairs and sorting the keys and the values independently, and hoping to get a meaningful result which is absurd.

Re:Machine Learning by stevenbedrick on Wednesday May 17, 2017 @12:07PM (#14359829)
That is an excellent analogy.
Re:Machine Learning by abandonliberty on Thursday May 18, 2017 @12:00AM (#14364716)
OMG. This is a data scientist trap. They can't help themselves! It's actually a great question to test fundamental understanding and ability to explain.
Re:Machine Learning by abandonliberty on Wednesday May 17, 2017 @11:53PM (#14364699)
Meanwhile I don't know how to get enough votes on my new stack exchange account to enable commenting.
Re:Machine Learning by v4nn4 on Wednesday May 17, 2017 @07:03AM (#14357285)
Wtf wow
Re:Machine Learning by nirmalpatel on Wednesday May 17, 2017 @09:30AM (#14358190)
Is this a joke?