There are several EAs based on AI methods in the CodeBase. There are also many Articles and Forum topics on AI, Neural Networks, Machine Learning etc., which include EA source code.
So, before posting your request, please use the website's Search function to find the related information.
Maybe someone knows some AI experts, or maybe someone has the source code. I will be grateful if you share.
Hello . Check these out in the sequence they are shared
(Perceptron JS example) https://www.youtube.com/watch?v=ntKn5TPHHAk
(Math for ML quick course) https://www.youtube.com/watch?v=T73ldK46JqE&list=PLiiljHvN6z1_o1ztXTKWPrShrMrBLo5P3
(MiT , Neural Nets Intro) https://www.youtube.com/watch?v=uXt8qF2Zzfo
(3B1B , Neural Nets Intro) https://www.youtube.com/watch?v=aircAruvnKk
(The creator himself course) https://www.youtube.com/watch?v=cbeTc-Urqak&list=PLoRl3Ht4JOcdU872GhiYWf6jwrk_SNhz9
(Neural network from scratch in Python) https://www.youtube.com/watch?v=Wo5dMEP_BbI
(Article about matrix and vector operations on mql5) https://www.mql5.com/en/articles/10922
(Back propagation solved example) https://www.youtube.com/watch?v=ZCzcmn6JxL4
(Article about programming a nn on mql5) https://www.mql5.com/en/articles/5486
(SVM MiT intro) https://www.youtube.com/watch?v=_PwhiWxHK8o
(SVM mql5) https://www.mql5.com/en/articles/584
Fun stuff :
https://www.mql5.com/en/articles/55
https://www.mql5.com/en/articles/8122
https://www.mql5.com/en/articles/11602
Back propagation will give you the hardest time , but imagine that you are trying to see the problem from a different multi dimensional perspective so with each layer you are going through an extra dimension in the hyperspace of what you are trying to solve . That is why you need activation functions to be differentiable . Keep in mind all these courses assume you are too familiar with mathematics or too familiar with the language they use .
Note , the editor f*ed up the playlist links , if you see a youtube link copy the entire line as it has the playlist it is in.
I've also got a RIdiculous Neural Net (RINN i call it) which uses the strategy tester's (beautiful) built in genetic algorithm to optimize itself .
If you want to see a simple structure (feed forward only) now , and have something to laugh at when you learn check it out
(note this is definately not the way to go about it , just for fun)![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use