Machine learning in trading: theory, models, practice and algo-trading - page 3379
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
Unstated terms are evil))))))
Once again, the fitness function is a method of model estimation. The robustness of the model will depend on which fitness function is chosen for robustness estimation.
Robustness can also be viewed as an estimate of the model. Robustness is the evaluation of the model for its ability to work on new data, i.e., it is the fitness function to be maximised.
I don't think these are very complex concepts that are difficult to comprehend, it is all the more surprising how many misconceptions there are on this topic.
Assessments, fitness function, can be tiered, with each tier controlling separate metrics. But for some reason, many people perceive the fitness function as something summarised, 'on top'.
Does it bother you that in response to my code of an example of a fitness function, and there are hundreds of them, you have a bunch of words again?
And also no one is confused as an optimisation algorithm top-3 or even top-1 in the world, generally recognised and well-known such as PSO he has at the end of the rating, and some know-names about which no one has heard of such as grey wolves, weeds, etc. He has leaders)).
Have you? Do your habitual patterns break when looking at tests?
Do something similar yourself, write a code in MQL5 of the most famous algorithms, describe and tell how your code works, and then post the results of the tests, so that anyone can reproduce these tests.
Then we'll talk about what questions anyone should ask and why.
For now, you are just idle chatter without grounds.
I don't take this rating as a rating because it is not clear by whom it is written, just copied algorithms in mql and fitted to a few functions. And in reality the shapes of the surfaces being optimised are very varied and there can be nuances.
Either his algorithms were implemented incorrectly or the experiment was performed incorrectly or both
In MO, the fitness function is used to train the model (parameter selection) through optimisation. A metric(s) is used to evaluate the resulting model. Often the metric does NOT match the fitness function. From a mathematical point of view, this means that the MO solves a MULTICRITERIAL optimisation problem rather than the usual single-criteria one.
Another significant difference from conventional optimisation is the frequent absence of a fixed set of optimisation prameters. Even for a regular tree this is already the case. From a mathematical point of view, this leads to an optimisation problem in FUNCTIONAL space, instead of the usual one in numerical space.
Both of these points make MO problems irreducible to conventional optimisation.