Discussing the article: "Neural Networks in Trading: Multi-Task Learning Based on the ResNeXt Model"

 

Check out the new article: Neural Networks in Trading: Multi-Task Learning Based on the ResNeXt Model.

A multi-task learning framework based on ResNeXt optimizes the analysis of financial data, taking into account its high dimensionality, nonlinearity, and time dependencies. The use of group convolution and specialized heads allows the model to effectively extract key features from the input data.

Among modern convolutional architectures, there's the one that stands out: ResNeXt introduced in "Aggregated Residual Transformations for Deep Neural Networks". ResNeXt is capable of capturing both local and global dependencies and effectively handling multidimensional data while reducing computational complexity through grouped convolutions.

A key area of financial analysis using deep learning is multi-task learning (MTL). This approach allows simultaneous solutions to multiple related tasks, improving model accuracy and generalization capability. Unlike classical approaches where each model addresses a single task, MTL leverages shared data representations, making the model more robust to market fluctuations and enhancing the training process. This approach is particularly valuable for market trend forecasting, risk assessment, and asset valuation, as financial markets are dynamic and influenced by numerous factors.

The study "Collaborative Optimization in Financial Data Mining Through Deep Learning and ResNeXt" introduced a framework for integrating the ResNeXt architecture into multi-task models. This solution opens new possibilities for processing time series, identifying spatiotemporal patterns, and generating accurate forecasts. ResNeXt's grouped convolutions and residual blocks accelerate training and reduce the risk of losing critical features, making this method especially relevant for financial analysis.


Author: Dmitriy Gizlyk

 

Dmitry, you have a huge number of articles on neural networks.

You prefer to earn money from writing articles rather than from trading.

Does it turn out that it is impossible to make money with a neural network?

 
I would like to know that too. Are there any trading EAs derived from this model?
 

Brothers in algocoding, very many programmers here, if not most of them, are studying and developing new technologies for themselves, and "yet" not making money on them.

After all, this is a forum of developers, not traders, mostly. Although there are successful traders. But we will never know about it.

 
Edgar Akhmadeev #:

Brothers in algocoding, very many programmers here, if not most of them, are studying and developing new technologies for themselves, and "yet" not making money on them.

After all, this is a forum of developers, not traders, mostly. Although there are successful traders. But we will never know about it.

From my experience, traders who have something really useful to share, never share anything. 
 
Alain Verleyen #:
In my experience, traders who can share something really useful never share anything.

Yes, they know (like me since 1998) that a working strategy quickly stops working after distribution.

That's why forum programmers share individual solutions, while a working (profitable) strategy has never been published. Or sold.

 
Edgar Akhmadeev #:

Yes, they know (as I have since 1998) that a strategy that works quickly stops working once it is disseminated.

That's why forum programmers share individual solutions, and a working (profitable) strategy has never been published. Or sold.

and the need to transfer funds between countries doesn't count anymore?)

How can you be such a system?

A trading robot will always work if you buy on a pullback, the question is where is the pullback?

 
I've seen the translation, I'm definitely not translatable.
 
lynxntech #:
I've seen the translation, I'm definitely not translatable

I have to admit, I wasn't smart enough to understand the original.

"I've been talking to myself all night, and they didn't understand me!" (Zhvanetsky

 
Edgar Akhmadeev #:
Yes, they know (as I have since 1998) that a strategy that works quickly ceases to work once it is disseminated.

This applies to exchanges with limited liquidity, it does not apply to forex, there is enough liquidity there for everyone

P.S. I remembered Mikhail, he has a system of hedging on the Moscow Exchange, he shared it and it works, and it should work in the future. Everything depends on personal capital, and there is nothing to do there with 100 dollars.

Here, everyone is looking for a system for a hundred quid, and profitability of 10% per day. That's why such results of searches.