Machine learning in trading: theory, models, practice and algo-trading - page 599

New comment
 
Yuriy Asaulenko:
I have no idea.) But it's not like that.

:D

You know what the thing is... I take the top products from the market, yes, for the sake of interest... I run them in the tester... for example a beaver. And I'm amazed at how primitive it is... But if you tweak it a little bit, it could run for a year or more.

It's not 100% per month... but it does 10%.

so you just need to pick up the key

 
Ivan Negreshniy:

Firstly - not exactly a fitting, since it is proved that committees work better, it is explained, for example, on opposite, in sign, deviations of separate models, leading to increase of accuracy. Secondly, the proposed tandem of NS is a single, self-adjusting model, which has also proven to be positive in reinforcement learning methods.


interesting, I haven't seen a description of such tandems anywhere... I'll have to look it up

 
Maxim Dmitrievsky:

:D

You know what the thing is... I take the top products from the market, yes, for the sake of interest... I run them in the tester... for example a beaver. And I'm amazed at how primitive it is... But if you tweak it a little bit, it could run for a year or more.

It's not 100% per month... but it does 10%.

so it's just a matter of picking up the key.

Again, picking... Imho, trying too many options is not the method.

Feedback is a very complicated thing. Thousands of monographs must be written about it. Theories such as automatic control systems are based entirely on feedback. I can say that nothing can be solved there using brute force.

 
Yuriy Asaulenko:

To pick up again... Imho, overkill is not the method.

Feedback is a very complicated thing. There must be thousands of monographs written about it. Such theories as Systems of automatic regulation (control) are entirely based on feedback. I can tell you that nothing can be solved there by trying.


Only an experiment will tell.

 
Maxim Dmitrievsky:

only an experiment will tell.

I think you should read Wiener's Cybernetics, or Control and Communication in Animals and Machines. The book is old, 1948. But, I suppose, in terms of outlook and new thoughts, it can give you a lot.

Although, maybe, it is not necessary to fill up your head.

 
Maxim Dmitrievsky:

Interesting, I haven't seen a description of such tandems anywhere... I'll have to look it up

Look it up, but the problem is that there is not much fresh, systemic information about working with low-level NS structure at scale level, since our researchers rarely even go down to backend, like TensorFlow, mostly everyone is spinning above or at Theano, Keras, Torch or unfading R level.
 
Ivan Negreshniy:
Look it up, but the problem is that there is not much fresh, systemic information on the web about working with low-level NS structure at scale level, as our researchers rarely even go down to backend, like TensorFlow, mostly everyone is spinning above or at Theano, Keras, Torch or unfading R level.
You just didn't look hard enough. Do you really need links or should I explain what is already written everywhere?
I do not know what you have difficulties:
  • Scientists make a hypothesis;
  • Investigate the question, prepare a mathematical basis;
  • They write an article/research result;
  • Enthusiast implements it based on well-known frameworks;
  • Anyone wants to use the ready-made tools in their work. Profit!

What's wrong with not digging into C++ code optimization, but just writing high-level code?

PS. Sorry, burned =)))

 
Yuriy Asaulenko:

I think you should read Wiener's Cybernetics, or Control and Communication in Animals and Machines. The book is old, 1948. But, I suppose, in terms of outlook and new thoughts, it can give you a lot.

Maybe I shouldn't overthink it, though.


Yes, I'll read it, thanks. Hope Stalinism is not too destructive influence.

 
Aleksey Terentev:
You searched poorly. Do you really need to get references, or can you explain what is already written everywhere?
I do not know what you have difficulties:
  • Scientists make a hypothesis;
  • Research the question, prepare a mathematical basis;
  • They write an article/research result;
  • Enthusiast implements it based on well-known frameworks;
  • Anyone wants to use the ready-made tools in their work. Profit!

What's wrong with not digging into C++ code optimization, but just writing high-level code?

PS. Sorry, I'm burning up =)))

Thank you, I'm fine, while it seems that you have difficulties, since you hang out in this thread and instead of a constructive discussion of questions, you interfere with other people's answers - yes, really, give your links, argue, explain, if there is anything and write at any level your code, but not suckers moral teachings.
 
Ivan Negreshniy:
Thank you, I'm fine, while you seem to have difficulties, since you hang out in this thread and instead of constructive consideration of questions, interfere in other people's answers - yes indeed, give your references, argue, explain, if there is anything and write at any level your code, but not suckers moral teachings.

Yes, I'm having difficulty. It's hard for me to understand why it's so hard for some people to get off their heels and work hard.
Yes, I hang out in this thread and interfere with other people's answers instead of considering questions constructively.
__Let's have a constructive discussion about deep learning? Using python? Learning with a teacher on good signals?
__I have no one to discuss with. And you just said you can't find anything. Well, that's not the case.
Yes, I'm giving a sucker's lecture. After all, my opponent himself used veiled mockery, confused terms, and made a couple of logical errors.

And I also apologized in advance, because I wrote in my feelings.

And also I offer you the help in knowledge of principles of work of tools for creation of neural networks. Without any irony or sarcasm.

1...592593594595596597598599600601602603604605606...3629
New comment