Algorithm Optimisation Championship. - page 6
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
Fewer FF starts are better, that's the point. This can be tricky.
You don't need to limit the algorithm, let it count for itself. Either it will decide to stop on its own, or it will be forced to stop. The algorithm doesn't need to know how many runs is the ceiling - no one will know the ceiling. There will be no disqualifications. As the algorithm was able to do, the problem will be solved.
Better for whom? If the participant's algorithm determines that the result is satisfactory, it can discontinue the task.
There's still the need to ensure that the algorithm can be interrupted by the checker. We used to talk about limiting the number of ffs calls. Now there are new ideas.
There is no way to make interruptions.
We don't need to complicate things. We should allow participants to be creative. Make a limit on the number of ffs calls and that's it.
Better for whom? If the participant's algorithm thinks the result is satisfactory, it can abort, its business.
There's still the need to ensure that the algorithm can be interrupted. Previously there was talk about limiting the number of ff calls. Now there are new ideas.
There is no way to do interrupts.
There is no need to complicate things. We should allow participants to be creative. Make a limit on the number of ffs calls and that's it.
Better in terms of getting a higher rating in the participants table. Knowledge of the maximum allowable ceiling of runs can be used and make runs much less than the ceiling, thereby increasing the chances of the algorithm in the table among the participants.
Everything will work out. Nothing gets complicated.
Better in terms of obtaining a higher ranking in the participants' table. Knowledge of the maximum allowable launch ceiling can be used and make launches much lower than the ceiling, thereby increasing the chances of the algorithm in the table among the participants.
Everything will work out. Nothing gets complicated.
You don't have to limit it, just count the calls. But if the search goes too long, just delete the script from the chart, and the participant is considered to have missed the whole time. But only if you are stuck for a very long time. Aren't you going to interrupt and watch the result?
It's simpler, much simpler.
The competitors surrendered the algorithms at the start of the championship. That's it, they can no longer influence the results.
Then a ceiling of FF starts is adopted by public opinion. Tests are passed. The algorithm counts the FF as many times as it wants. If it has made more runs than the limit, the script stops.
It is elementary.
Where's the chance of that? Few challenges - bad result. Hope for randomness or something?
Goal: the best result for the least number of runs at the highest intrinsic speed(rule 3). Competitors will be ranked according to these three criteria. Improving any of these criteria is an upgrade in the table. Reducing the number of FF runs is the shortest way up the table.
Randomness is not the worst search option, I assure you. I recommend those who don't particularly want to bother with algorithms to apply just the HGC.
It's simpler, much simpler.
The competitors surrendered the algorithms at the start of the championship. That's it, they can no longer influence the results.
Then a ceiling of FF starts is adopted by public opinion. Tests are passed. The algorithm counts the FF as many times as it wants. If it has made more runs than the limit, the script stops.
It is elementary.
It can be written in rules of participation - in function of the participant the maximum allowable quantity of calls is transferred and it should interrupt itself at achievement of this quantity.
There is no way to interrupt from the outside, without complicating the participant's function, which is actually what we are talking about.
It can be written in the rules of participation - the maximum allowed number of calls is passed to the participant's function and the participant must interrupt itself when this number is reached.
There is no way to interrupt from the outside, without complicating the participant's function, which is what the whole discussion is about.
You can do this - the number of ff calls allowed is defined - the main parameter.
A time limit is defined, e.g. 5 min or 10, if the search is not completed within this time, interrupt and do not look at anything. This is just in case of slow algorithms.
The result is shown by value.
How can this not be interrupted? The executing script (common to all) will be unloaded and that's it.