Machine learning in trading: theory, models, practice and algo-trading - page 1749

New comment
 
Maxim Dmitrievsky:

record a video with explanations, it's so unclear

I mean that the "weight" language of the network, gives it universality, and it can work with any types of data, (after converting them into "weight"). This is unification (reduction to a single system/form).
 

Went to choose a color for the lamp...

Don't tell me there's a mistake, I searched ... did not find ... I hope that I will not find it))

 
Aleksey Nikolayev:

A multidimensional function is an ordinary mathematical function whose definition area is a multidimensional space. In the case of NS it is the feature space.

I want to ask you as a mathematician - did you study mathematics at school at all?)

I studied it. But, what does the essence of the NS device have to do with it?
The space, the signs, the search, are not the point, but part of the work. As individual words in a sentence, they do not convey the meaning and do not answer the question.

 
Aleksey Nikolayev:

...

Describe the essence of NS device, if you understand it. Terms from mathematics and physics, by themselves, are not enough to describe the mechanism of operation and the device.

I remember studying the structure of car systems from books. So, the systems were described thoroughly, thanks to the concise formulation of the essence.
 
Tag Konow:
I mean that the "weight" language of the network, gives it universality, and it can work with any data types, (after converting them into "weight"). And this is unification (reduction to a single system/form).

the data is not converted into weight, but each instance is weighted, and each input begins to affect the final result differently

An artificial neuron is a purely mathematical object, it doesn't make any physical sense

so you have to understand it through mathematics

Download Heikin's book "neural networks full course".
 
mytarmailS:

Went to choose a color for the lamp...

Don't tell me there's a mistake, I searched ... did not find ... I hope that I will not find it))

not much of a story

what's the method?

 
Konow's retrigger:
Learned. But, what does this have to do with the essence of the NS device?

From the mathematical point of view, any NS is a parametric family of multidimensional functions. A set of weights are the parameters that define this parametrization. Learning - defining specific values of weights, which corresponds to the selection of a particular multidimensional mapping from the original parametric family.

 
Maxim Dmitrievsky:

the data is not converted into weight, but each instance is weighted, with each input starting to affect the final result differently

The artificial neuron is a purely mathematical object, it doesn't make any physical sense

so you need to understand it through mathematics

download the book Haykin "neural networks full course"
Downloaded it, thanks.
 
Maxim Dmitrievsky:

little history

what is the method?

1.spectral analysis

2.adaptive systems

3.forecasting

4.MO

5. And two dummies at the end of "DEMA" )))) for smoothing the signal

 
Konow's tag:
I haven't found an independent definition of "multivariate function". There is a "distribution function" of probability theory, and within it is a view of "multivariate distribution functions," but there is no mention of MO technology.

Obviously, multidimensional functions, if they have anything to do with NS, are far from its essence. Probably something to do with the implementation of some technical nuances. I, on the other hand, am trying to understand the essence.

x=y+z, although yes, it is strange that definitions, if the arguments are more than 1 then it ... not on the fly.

1...174217431744174517461747174817491750175117521753175417551756...3495
New comment
Reason: