Machine learning in trading: theory, models, practice and algo-trading - page 1675

New comment
 
Aleksey Nikolayev:

We are talking about zero tolerance to it, declared by the administration - they promise a lifetime ban. But for some reason this rule does not apply to some magicians.

Knock louder, with your head, there's no shit there anyway.

 
Vizard_:

Aleshenka, you're a fucking patient man.)
While the normal guys are scrubbing kitchens, you're running around with Ran.
You don't run a decompile in a protected sandbox, you snot all over the thread.
Look for the half-wit, this dcuompil is in the public domain on some forum,
I don't remember which one. Don't say anything about ml, you don't know shit, especially about
especially about the non-stationarity. Ten years ago it was explained to the FA shithead,
but he didn't get it. Open up a branch with your fucking briefcases and talk about it.
Fucking miners, for Christ's sake))) hilarious...

The magician broke up, bring a new one.)

 
Vizard_:

How many times have we coddled you, you ungrateful bastard?)
Kesha's a good-looking guy, but what are you going to eat tomorrow?
I don't know what you're gonna eat tomorrow...

Here comes the heat. Come on... it's my turn to laugh. You must have flushed yourself out if you're so nervous. I can smell it on you.... You and i are of the same blood. You... not me.... :-)
 
Mihail Marchukajtes:
Enlighten me, will you?
Google the 4 types of machine learning. Regression, classification, clustering, anomaly detection. They may have invented something else with reinforcement.
 
Vizard_:

Knock louder, louder with your head, there's no bullshit there anyway.

I don't see the point. I guess you're a virtual of one of the knocking receivers.

 
Aleksey Nikolayev:

The magician broke off, bring a new one)

He's all torn up today)) Trickster, Ale, give me presents.
 
Maxim Dmitrievsky:
Google 4 types of machine learning. Regression, classification, clustering, anomaly detection. They may have invented something else with reinforcement there.

Clustering is a partitioning of input objects into classes, which may be many. Classification is a partitioning into two classes, well in our case three. So clustering is a special case of classification. And this is without wikipedia, just... off the top of my head.

In general, there are two global trends in MO, prediction and classification. Whether you have examples of directions which it would be impossible to carry to this or that group???? Very much would like to listen....

 
I'd like to add, "ETH and don't f...di," but I'll wait for your answer.... Maxim...
 
Mihail Marchukajtes:

Clustering is a partitioning of input objects into classes, which may be many. Classification is a partitioning into two classes, well in our case three. So clustering is a special case of classification. And this is without wikipedia, just... off the top of my head.

In general in MO there are two directions of global, this prediction and classification. Do you have any examples of directions which it would be impossible to refer to this or that group???? Very much it would be desirable to listen....

Now try through Wikipedia. What is the difference between learning with a teacher and learning without a teacher
 
Maxim Dmitrievsky:
Now try Wikipedia. What is the difference between learning with a teacher and learning without a teacher?
The difference is that in the first case we try to assign the input vector to this or that class from the target class. In the second case we try to divide vectors into groups according to the distance of these vectors (conditionally). Just methods are different but the essence is the same. That is why clustering is a subspecies of classification. Questions???
1...166816691670167116721673167416751676167716781679168016811682...3629
New comment