Machine learning in trading: theory, models, practice and algo-trading - page 922
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Maxim, what do you use as predictors? If it is not a secret, of course.
Now just price series, not transformed in any way for RF
for NS you can take normalized increments
But I don't have the Grail, I'm not an authority here. Aleshenka here hasCan you briefly state the essence of the binge? The language barrier prevents me from understanding...
When I get to the scaffolding according to your article, I'll ask a lot of questions.
the gist of the hilarity in a nutshell. The Chinese always cheat, and this is all my ideas from the articles.
removes noise from predictors via fuzzy rules, then outputs are fed to inputs of recurrent network, which is trained via reinforcement
but, The financial problem is highly complicated and it is hard to manually set up the fuzzy membership functions
Fig. 2. Overview of fuzzy DRNNs for robust feature learning and self-taught trading.
We prefer to learn the fuzzy membership functions directly from the experiences.Therefore, we prefer to directly learn the membership functions and this idea will be detailed in Section IV.
In a nutshell, the gist of it. The Chinese always lambast everything, and here I have all my ideas from the articles
removes noise from predictors through fuzzy rules, then outputs are fed to inputs of recurrent network, which is trained through reinforcement
But, The financial problem is highly complicated and it is hard to manually set up the fuzzy membership functions
Fig. 2. Overview of fuzzy DRNNs for robust feature learning and self-taught trading.
We prefer to learn the fuzzy membership functions directly from the experiences.Therefore, we prefer to directly learn the membership functions and this idea will be detailed in Section IV.
So the idea is really good!
So the idea is really good!
It's a common idea, an everyday idea. Every housewife every now and then is thinking about such neural network architectures
I'm interested in details, but you can't find them anywhere until you think them up yourself.
https://github.com/Roffild/RoffildLibrary
Discussion (I hope they don't kill the thread again):
https://www.mql5.com/ru/forum/245373
Never mind :)
Now vps on google... At your suggestion. Micro. It seems to be fine, but I have not yet reached the mode.
Some other freebies on Microsoft's azure) - have not check it out yet, since I do not need it.
Microsoft also has free stuff like that on azure) I haven't looked at it yet, because I don't need it.