NATURAL INTELLIGENCE as the basis of a trading system - page 95

 
xmaster писал (а) >>

I'm sorry, I'm not good with terms. What is a DC?

In Trader Workstation from Interactive Brokers.

 
Integer писал (а) >>

What DC do you get like that - 20 trades in 1 (one) minute.

Integer, these are futures, E-mini, traded on the Chicago Stock Exchange CBOE, the speeds there are enormous - this is the fastest trading, it is done through a stock broker, it has nothing to do with forex. Stop thinking in terms of brokerage companies, it's a Soviet invention.

 
Yurixx писал (а) >>

Integer, these are futures, E-mini, traded on the Chicago Stock Exchange CBOE, the speeds there are enormous - it is the fastest trading, it is done through a stock broker, it has nothing to do with forex. Stop thinking in terms of DCs, that's a Soviet fiction.

You can also think in terms of your own places on the exchange, in terms of exchanges, countries, continents, planets, star systems, galaxies. But still, whoever I asked this question to - got it right.

 
Integer писал (а) >>

Good reasons.

Very convincing!

I'll try, too: in the attachment is a picture of intelligent programmes.

I hope to shake your conviction :)

Files:
081108.zip  344 kb
 
Not shaken. You wrote about one thing, but you imply something else. The application of neural networks is one thing (demonstrated by you), but the invention (or development) of fundamentally new methods of their functioning (not application!) is an entirely different topic.
 
TheXpert писал (а) >>
If you've made a breakthrough in neural network technology, I think it's worth... for at least a PhD.

Dear Expert!

I have not made any breakthrough in neural networks as I do not formalise living things (sorry - idiocy) - that is the lot of mathematicians, who see the output response of a living thing as a result of some transfer function of the input signal. Neural networks allow them to find this and many other completely meaningless functions much faster. So they're playing around,

Despite the fact that 50 years ago P.K. Anokhin explained senselessness of such an approach to modelling (even the most primitive) living things.

However, the corrupting influence of the West and domestic pillars of 'cybernetics' (mostly mathematicians) managed to successfully bury this science for many years. The young seem to be carrying on the glorious tradition of their fathers!

But in understanding the structure of living beings, the principles of their functioning, mechanisms, machines and everything else a breakthrough has really happened...! But somehow imperceptibly :(

And it was P.K. Anokhin who made it approximately 50 years ago! I simply studied his works attentively and translated what I understood into computer-understandable language :)

By the way, P.K.Anokhin is an academician, a Russian scientist of genius. So your "doctoral" grades are obviously understated even by Soviet standards :)

Yes, by the way, my programs almost do not use mathematics (well, if only at the level of arithmetic). And for comparison with neural networks the memory size of what I call brain is 200-500Mb

 
Integer писал (а) >>
No, it's not. You wrote about one thing, but it turns out you mean something else. Application of neural networks is one thing (demonstrated by you), while invention (or development) of fundamentally new methods of their functioning (not application!) is an entirely different topic.

So it's not meant to be... Wrong assessment...

Sorry to bother you :)

 
Pterovich_I писал (а) >>

I'll try it too: in the attachment is a picture of the sensible programmes.

In the attachment is a picture of the interface of a program which gives signals or forecasts (I don't know what you call it) for ES-traders.

From which part of this picture is it obvious that the program is reasonable ?

I agree, the results are convincing. But we are not talking about results, but about the reasonableness of the programme.

A human being is a reasonable creature, by the way. Nevertheless, successful traders among humans are very rare.

So it's worth explaining at least what meaning you put in the term "reasonableness" of your program.

However, if your purpose is to promote your website, you should not bother, it is not the audience.

 
Yurixx писал (а) >>

Attached is a picture of the interface of a programme that gives signals or forecasts (I don't know what you call it) to ES traders.

From which part of this picture it is clear that the program is reasonable ?

I agree, the results are convincing. But we are not talking about results, but about the reasonableness of the programme.

A human being is a reasonable creature, by the way. Nevertheless, successful traders among humans are very rare.

So it's worth explaining at least what meaning you put in the term "reasonableness" of your program.

However, if your goal is to promote your website, then don't bother, not the right audience.

To quote myself (don't know how to give a link :-) :

I call my systems Intelligent as opposed to Intelligent because so far nobody in the world of cybernetics has given an objective definition of the term "intelligence" itself, based on objective laws of nature, rather than on a descriptive - subjective level.

I believe that if a programme can learn, with or without a teacher, to perform rational and result-oriented actions, it can be called intelligent, just like a human.

So, what you have seen in the picture is just a confirmation of the theory in practice.

My systems are learning to trade and as you can see they do it quite successfully.

In this case, I only construct a "genotype" and improve it following the results of "natural selection". Everything is just like in nature.

I got to your forum by accident and was delighted to see attempts at philosophical analysis of life and the opportunity to communicate with normal people in their native language :) many thanks for that.

Besides I don't like NS for a long time and tried to draw your attention to achievements of compatriots and non-important figures.

Yes, by the way, even dull-headed Americans from such organizations as DARPA are already asking for something different from NS. Even there have already realized futility of this idea.

It has become a shame for Russia :(.

I would like to see the achievements of Neural Networks (because this topic was touched), although I know the result beforehand :-).

And I do not "promote" the service - I make people rich :-).

However, you can not force ...

So, let's talk about reasonable systems.

 
Pterovich_I писал (а) >> I would like to see the achievements of Neural Networks ( as this topic has been touched upon) although I know the outcome beforehand:-)

https://www.mql5.com/ru/users/Better