Algorithm Optimisation Championship. - page 102
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
By now you should have at least some algorithm, but you don't have one. I told you not to stick to functions, no analytical constructs and solutions will help you, because you have no knowledge about the contents of FF. I cited a thousand times the example of a staff optimizer, he knows nothing about the tasks of users, nothing, and yet successfully solves the optimization tasks.
Create a universal algorithm, find it on the Internet, use random search, whatever, but you should already have a ready-made algorithm. How are you going to compete in something you don't already have?
SZZ Anyone who has any algorithm at all is still welcome to join the competition, since apparently I'm really the only one left. But in any case it doesn't change anything, my algorithm and test bench will be shown within a couple of days, no matter what.
What nonsense!
DON'T GIVE ME ANY ADVICE.
I'm not asking you to explain to me how the algorithm should work.
I'll work out my own algorithm and its versatility etc..
All I need is to understand the rules and the task at hand.
The time to prepare the algorithm can start after the FF library has been put up.
The impression is that you want to get away with a fair contest... Why are you acting like this?
I see. You've decided to tighten the screws. No preparation time, no compromise, no joint decision making.
If I never write my algorithm and don't fight you, will you still consider yourself the winner?
Doesn't such a victory look pathetic?
Think what you like, but the fact is that I have an algorithm and you don't. I can fight on any terms, and you only on those that you are willing to discuss. Draw your conclusions. You still have a couple of days, maybe you'll make it. Otherwise, you compete next time, than held hopefully not the last time and maybe look at my achievements someone wants to surpass them and give me a real challenge, not a bluff.
And what do you want to discuss? Speak specifically, ask specifically without "or maybe...", I'm running out of time, I need to provide the source code for the test bench today.
Think what you like, but the fact is that I have an algorithm and you don't. I can fight on any terms, and you only on those that you are willing to discuss. Draw your own conclusions. You still have a couple of days, maybe you'll make it. Otherwise, you compete next time, than held hopefully not the last time and maybe look at my achievements someone wants to surpass them and give me a real challenge, not a bluff.
And what do you want to discuss? Speak specifically, ask specifically without "or maybe...", I don't have much time, I have to provide the source code of the test bench today.
You have already been told everything.
If you are so afraid of the competition that you do everything to stay alone and call yourself a winner, of course, quickly tighten the screws without even giving me to take the run-up.Keep on laughing, I'll keep on laughing. You'll see your smile come off your face. :)
You have already been told everything.
If you are so afraid of competition that you do everything to stay alone and call yourself a winner, of course, quickly tighten the screws without even giving me the run-up.Peter, please be sensible! The championship is possible only among trained participants, and you have no algorithm. Do you understand? There is no algorithm. What do you want to compete with?
I've been telling for two months in a row: do not tie yourselves to particular function formulas, but use a universal algorithm, just as real trading problems are solved in real life. We are not competing with matcads and mapples that solve problems analytically, we are competing in solving problems with an unknown analytical or any other formula, the same problems that are solved every day by MT's staff optimizer.
I urge again! All who have READY optimization algorithms - join us. A couple or three days later we will conduct the Championship of Optimization Algorithms!
Keep on laughing, I'll keep on laughing. You'll see your smile come off your face. :)
Competition is meaningful when algorithms that know how to find extrema are competing! Your rules don't contain a paragraph checking algorithms for the ability to find an extremum. What is there to compete with, if the correctness of the algorithms presented is not known?
There are two ways to check the correctness of the work:
1. Theoretically, i.e. using mathematical methods to check the convergence of the formula of the algorithm. (in our case is not possible, because we do not know the formula of the algorithm);
2. Practically, i.e. checking the algorithm's ability to find extrema on known data with a known global extremum (what is needed).
This is trivial.
After passing such a check, the algorithms are allowed into the Championship.
When the algorithms have been submitted to the referee, the latter
- forms the FF. He is not forbidden to change the values of the FF parameters. Then all the tinkerers will be where? Exactly in the arse;
- gets the extreme value, by a complete enumeration;
- publishes the obtained value to the public;
- runs the participants' algorithms.
Everything!
You've missed a lot, so you're reasoning on the wrong plane.
And by the way, do you have a Finished Optimisation Algorithm?
Peter, please be sensible! The championship is possible only among trained participants, and you have no algorithm. Do you understand? There is no algorithm. What do you want to compete with?
I've been telling for two months in a row: do not tie yourselves to particular function formulas, but use a universal algorithm, just as real trading problems are solved in real life. We are not competing with matcads and mapples that solve problems analytically, we are competing in solving problems with an unknown analytical formula, such problems that are solved daily by MT's optimizer.
I urge again! All who have READY optimization algorithms - join us. After a couple - troika days we will organize Championship of Optimization Algorithms!
You're an actor, dear boy...))).
So, you will not compete with me, because I do not have an algorithm.
And how can it be if you haven't put up the FF library yet?
Your "I keep telling you to make a universal algorithm" looks like a childish "buy an elephant".
Peter, please be sensible! The championship is possible only among trained participants, and you have no algorithm. Do you understand? There is no algorithm. What do you want to compete with?
I've been telling for two months in a row: do not tie yourselves to particular function formulas, but use a universal algorithm, just as real trading problems are solved in real life. We are not competing with matcads and mapples that solve problems analytically, we are competing in solving problems with an unknown analytical or any other formula, the same problems that are solved every day by MT's staff optimizer.
I urge again! All who have READY optimization algorithms - join us. A couple of days later we will organize Algorithm Optimization Championship!
I don't have time to develop my own algorithm. I found a simple classical method of search for extrema on the Internet. I am trying to program it.
This method specifies the number of variables of the analytic function. Will it be strictly defined at the championship?
You're an actor, monsieur...))
So, you won't compete with me, because I don't have an algorithm yet.
And how can there be one if you haven't put up the FF library yet?
Your "I tell you to make a universal algorithm" is like a child's "buy an elephant".
You still haven't figured it out. That's too bad.
What difference does it make to you if you have or don't have the FF library? You must have an optimization algorithm that can solve problems with unknown content and unknown formula. Any optimization task must be within its power. This is what the championship is designed for, it has been said a million times.
Why am I not afraid of any FFs and you are? Why can I solve any optimization problems even without knowing the innards of the FF, and you still can't? I told you, everything goes as planned, we do not change anything else, either you perform all steps synchronously with me or not - it is up to you to decide.
Ask what is not clear, specific questions on solving problems with an unknown formula, I will answer, but the championship will run its course no matter what. If you do not have time - well, it happens, you're not the first or last in such situations, you'll have time next time, when you look at the test bed and chase my algorithm, and those who may still have time to join the competition.
Take the Alglib algorithm from the codebase, it can solve the championship problems if you haven't been able to prepare your own algorithm so far. It will solve the problem without unnecessary questions or demands to change the championship conditions. Take it, you might even win with it this time.
I no longer have time to develop my own algorithm. I found a simple classic method for finding extrema on the internet. I'm trying to program it.
The number of variables is defined in this method. Will this be strictly defined in the championship?
There! Respect, respect! Whoever is looking for it will always find it. And if he doesn't like what he finds, he'll remake it for himself! - I'm not being sarcastic, I really respect it.
You may find out the number of FF variables by calling the corresponding function
int GetParamCount ();
Although the number of variables is known - they are 500, but still I recommend you not to stick to the number 500, use the function as shown above.