Machine learning in trading: theory, models, practice and algo-trading - page 1979

 
Maxim Dmitrievsky:

doing full logging, it will tell you what it's doing

through this you can then understand what to improve

In the full log is like a forest. Searching is difficult. Although will be sure that nothing is missing.

 
Maxim Dmitrievsky:

What are you doing there? Have you figured out how that thing works?

I just got home, I've been resting all week.

 
mytarmailS:

What's up? Did you figure out how that thing works yet?

I just got home, I've been resting all week ...

in general terms, I'm going to write an analog, I'm studying the packets.

i think it could be dramatically improved with new architectures

 
Maxim Dmitrievsky:

in general terms, I'm going to write an analogue, I'm studying the packets

I think it can be dramatically improved by new architectures

What's the basic scheme of operation? What's learning where and how, what it remembers, etc...

The author of the algorithm is clearly not a moron, and I'm sure that such letters as RNN, ltsm, he knows exactly how to work, but for some reason did not apply them.

It is necessary to clearly understand how things work, from the beginning ...

 

came across... they say it will be 4th generation networks

https://www.nature.com/articles/s41598-020-64878-5

Experimental Demonstration of Supervised Learning in Spiking Neural Networks with Phase-Change Memory Synapses
Experimental Demonstration of Supervised Learning in Spiking Neural Networks with Phase-Change Memory Synapses
  • 2020.05.15
  • Y. Lecun, Y. Bengio, G. Hinton,
  • www.nature.com
Spiking neural networks (SNN) are computational models inspired by the brain’s ability to naturally encode and process information in the time domain. The added temporal dimension is believed to render them more computationally efficient than the conventional artificial neural networks, though their full computational capabilities are yet to be...
 
Maxim Dmitrievsky:

came across... they say it will be 4th generation networks

https://www.nature.com/articles/s41598-020-64878-5

It will be expensive at first)

 
Who makes money from machine learning (besides trading) and how?
 
Evgeni Gavrilovi:
Who and how earns from machine learning (besides trading)?

Oh what an interesting question!!! Really where else can you apply MO for earning....

As an option betting on sports. I even made models for sweepstakes BUT there is one BUT, it is extremely difficult to obtain arrays of data in automatic mode, but in general it is very interesting thing to beat sweepstakes....

 
mytarmailS:

Well, what is the basic scheme of work? What where and how to learn, what it remembers and how, etc...

The author of the algorithm is clearly not a moron, and I am sure that such letters as RNN, ltsm, he knows exactly how to work, but for some reason did not apply ...

It is necessary to clearly understand what works, from the beginning ...

What is the circuit diagram?

I've only been in the woods before, I've never used LNs except for primitives. I'm getting into it now, little by little. The idea of memory and new architectures intrigues me, but don't get too carried away.

 
Maxim Dmitrievsky:

What's the circuit diagram?

I've only wandered in the woods before, I haven't used HH-ki except for the primitives. Now I'm plugging it in little by little. The idea with memory and new architectures intrigues, but without fanaticism.

How do they organize the memory?
Reason: