Machine learning in trading: theory, models, practice and algo-trading - page 1700

[Mihail Marchukajtes]

Fucking second moose. I'm going underground.... The market punished me for the booze :-(

[Реter Konow]

mytarmailS:

There is no problem creating such a network , and it will also recognize patterns no worse than a human, even better...

The funny thing is that the network will not make money on these patterns just as well as a person who recognizes them at once )) !!

The problem is in the patterns, not in the network and in the teacher who wants the network doesn't know what it is, but he thinks he does.

It's not about making money. It has nothing to do with it. The question is about using the tool to find fuzzy shapes and signal. And then how the TS uses it is secondary.

[mytarmailS]

Konow's tag:
A working model of human intelligence possessing, as it does, universality in terms of problem solving. "Absolute Solver".

It's kind of a fuzzy concept.

Artificial intelligence is a working model of human intelligence?

Which, like it, has the universality of problem solving?

How will AI (in your definition) solve this universal task if it's in a computer?

Your definition is lame.

Retag Konow:
It's not about making money. It has nothing to do with it at all. The question is to use the tool to find fuzzy shapes and signal them. And then how the TS uses it is secondary.

I wrote that you can

[Forester]

Aleksey Vyazmikin:

If so, that's not good enough.

So far my comparison experiment is not finished - I found a mistake in data preparation and had to put 8 samples again for training.

I think that by the end of the week there will be the result, which I will publish here.

And, I'll do my "enumerator" of conditional leaves on these predictors (by the method that I described earlier) - with the control of the results (different metric) at each step.

The entire market cannot be described by a single model, so I'm going to search for stable patterns by combining them in TS and I'm not worried if some part of the "space" remains undefined.

What's wrong with it? Catbust beats most of its competitors.
I, for example, did not like the idea of a symmetric tree in it. Obviously, 2 different nodes sharing the same predictor and the same level is not the best option. Unless the acceleration is up to 10 times.
Good thing they added 2 new more classic methods.

[Реter Konow]

mytarmailS:

It's kind of a fuzzy concept.

Artificial intelligence is a working model of human intelligence?

Which, like it, has the versatility to solve problems?

How will AI (in your definition) solve the problem, if it's in a computer?

Your definition is lame.

The question is also fuzzy. It's like asking "what is the World?")

In my definition, after setting a task with a hundred-meter run, the AI should ask you how it will do it if it's in a computer? )) There, that's the real AI.)))

That is, the universal solver doesn't go into a stupor from the impossibility of one solution, and finds an analog in the form of another solution. For example, it can draw an animation with a treadmill and a human.

[Mihail Marchukajtes]

In general, the engineer's task is to facilitate the work of the NS as much as possible and then it will have the internal potency for adequate solutions. Philosophically of course...

[mytarmailS]

ReTag Konow:
The question is also fuzzy. It's like asking "what is the World?")

That's exactly the question.

Okay I will not torture you, the point is that you confuse artificial intelligence with artificial intelligence.

Intellect is just a tool to solve a problem, it's part of the mind, but it's not the only part.

Artificial Intelligence has been around for a long time, but it's not what you think as you have the wrong concept of intelligence and hence the wrong concept of AI

[Реter Konow]

mytarmailS:

The question is exactly that.

The answer to the question is in the question itself. Artificial Intelligence. First, let's define what Intelligence is. Then, let's understand what it means to be artificial. The list of Intellect properties is endless. But to summarize, we can say that it is a universal solver, modeler, predictor, and so on... a logical "machine."

Okay. It is a super-complex system consisting of parameters that reflect a picture of the world.

[mytarmailS]

Tag Konow:
The answer is in the question itself. Artificial Intelligence. First, let's define what Intelligence is. Then, let's understand what it means to be artificial. The list of properties of Intellect is endless. But to summarize, we can say that it is a universal solver, modeler, predictor, and so on... a logical "machine."

Already identified.

read me, there are pages 4-5https://www.mql5.com/ru/forum/86386/page1681

[Mihail Marchukajtes]

mytarmailS:

Already identified.

read me, there are pages 4-5https://www.mql5.com/ru/forum/86386/page1681

I think I got into the discussion there, too.