Artificial Intelligence - page 14
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
biddick
I've heard about FURIA and read 2 or 3 white papers about it but I don't have any experience using it. It's based on RIPPER and the results aren't a lot better so I just added it to my "to do" list for later.
Anyone knows of EA's designed to Profit ...??
Just checking,
is it that all EA's ultimatley result in a loss over time?
Is there like an EA Junk yard where traders with new ideas can tinker an probably inprove the performance of current poor performers (that were once profitable).
It would be nice to see what an EA with a - 1 expectancy does to an account. this seems the likely effect of an EA designed to optimize itself and no longer can adapt to current conditions.
If you have your discarded EA and they can be optimized, please share same , one never knows fron where the next theory on relativity might emerge.
Peace.
for those who dont know what is NN, I hope this will help
a book ""
""
Publisher: Chapman & Hall/CRC
ISBN-10: 1584882298
ISBN-13: 978-1584882299business economics with Matlab
Welcome, MiHD.net - World's most generous file web hosting serviceThe original code for Artificial Intelligence EA can be found at Artificial Intellegence - MQL4 Code Base .
It uses a simple perceptron algorithm, I wonder if there is anyone who can replace the simple perceptron with a "Voted perceptron" algorithm that is found in this paper attached.
Need some help for this EA
i started playing with this AI FX ROBOT v1 i found somewhere
did some backtests, and found a good setting, but thing is, i want just one tp level i dont want it to close half position and move sl to BE like it does as standard, the file does not open in metaeditor can someone help me please.
Is anybody using this (some forward instead of back test results would be interesting)?
I eat GMO food...We are what we eat...so...I consider my intelligence already Artificial:)
By now everything is artificial. Was just interested if someone uses this one or not
Is anyone still active on this thread?
To train the Perceptron algorithm initially posted do we run several series and manually adjust the values of our variables?
Even if MT4 is too slow, wouldn't it be possible to make a part of code running aside to calculate optimization for these variables?
(I mean in principle, or if it is even useless that I think about implementing it).
Thanks!
Is anyone still active on this thread?
To train the Perceptron algorithm initially posted do we run several series and manually adjust the values of our variables?
Even if MT4 is too slow, wouldn't it be possible to make a part of code running aside to calculate optimization for these variables?
(I mean in principle, or if it is even useless that I think about implementing it).
Thanks!Hi DadeM,
This thread looks fairly inactive...
And I haven't tried the AI EA so I don't have any answers for your questions.
However there are some other threads here on Forex-TSD on Artificial Intelligence and Neural Nets that are more recent.
Also WinterSky has a thread on Statistical Analysis that crosses over to AI and NN...in one case using statistical programs called R and Rattle.
Two of the most recent advances in Neural Nets (for beginngers like me) are ENCOG and NEUROPH...
Both have good platforms with a relatively easy to use GUI...so people like me don't have to code.
We beginners can now build neural nets to play with...and visually see the connections and weights.
Neuroph may be the easier to start with...and has many good examples to learn from...
The downside is you have to learn to normalize your data before you use it in Neuroph.
Encog allows for more complex developments...and does auto-normalization for the data so you can let the program do it for you.
I have seen a few articles on connecting Encog and MT4...so it can be done...
Another downside is that support for both is somewhat lacking...and I would hope that starting a thread here for Neuroph and/or Encog may stimulate interest and help for us beginners.
As I said...I'm just starting to learn about neural nets...and Neuroph and Encog are a good place for others to start looking.
Hope this helps you,
Robert