Algorithm Optimisation Championship. - page 76
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
I'm here. Aukai out.
You can also be trusted, you're out of competition anyway, and it's not in your interest to peek.
There is also a candidate for referee, he knows and understands everything, but does not want to participate as he is afraid to be last.
I think he's smart enough to compile the FF, it's abolk.
Nah, I'm already under a lot of suspicion, and I need an ironclad alibi, even to myself, so just MQ - that's out of the question. But thanks for the vote of confidence!
Did you try the word problem? - How did it turn out?
Have you tried the text problem? - How did it turn out?
Haven't tried it yet, other things are distracting me. There's still plenty of time till the 11th )
To solve this problem, I would make a different FF, to make it smoother, or you can make a simpler, faster and more accurate algorithm. My algorithm is sharpened more for neural networks. But a task is a task, I don't know what the result will be, so I'm not going to tweak the algorithm, whatever comes out will be fine. I think I will find an hour over the weekend to tweak the algorithm to the task.
If you were trading, or developing some kind of expert for yourself, you would get more out of it than your competition.
Why spend so much energy on it.
It is also an optimisation task.
If you were trading, or developing some kind of expert for yourself, you would get more out of it than your competition.
Why waste so much energy on it.
It is also an optimization task.
And if everything is already developed, what to do, where to spend the energy?
p.s. We need to measure who has the fastest algorithm ))
Haven't tried it yet, other things are distracting. There's still plenty of time until the 11th )
To solve this problem, I would make a different FF to make it smoother, or you could make a simpler, faster and more accurate algorithm. My algorithm is sharpened more for neural networks. But a task is a task, I don't know what the result will be, so I'm not going to tweak the algorithm, whatever comes out will be fine. I think I'll find an hour over the weekend to tweak the algorithm for the task.
That's the thing about it, it's not smooth. And with a smooth FF any fool (algorithm) can do it. :)
Be sure to check your algorithm on this function, it should be able to solve such discrete functions. With neurons - yes, their FFs are more or less smooth, but quotes are discrete by nature, and the results of any TS are discrete, that's why the text problem is so important for training.
If you were trading, or developing some kind of expert for yourself, you would get more out of it than your competition.
Why waste so much energy on it.
It's also an optimisation task.
You may consider this task as a hobby of mine. Firstly, it is very interesting, and secondly, it is very useful in solving practical tasks.
Take climbers and ask them "why do you spend so much energy?" They will answer you that it is not only fresh air (which stimulates thinking activity), but also a training of all muscle groups, adrenaline and other delights.
And what makes you think that I do not develop trading robots and do not trade? - You are wrong.
Maybe you should sell EAs for 100500 roubles each and a few signals demonstrating "profitable" trading without SL. Then people would recognise that the marketer is in fact doing a good job! And if he hasn't got anything in the Market or in his signals, he'll organize championships out of necessity...
That's the thing about it, it's not smooth. And a smooth FF can be handled by any fool (algorithm). :)
Be sure to check your algorithm on this function, it should be able to solve such discrete functions. With neurons - yes, their FFs are more or less smooth, but quotes are discrete by nature, and the results of any TS are discrete, that's why the text problem is so important for training.
I'll try it, but I disagree about the quotes, it depends on which way you look at it, the quotes are very smooth.
Y=X^2 is an example of a smooth function. That is, any function F(x) is smooth if any arbitrarily small change in the value of X leads to a change in the value of F other than 0.
That's why any quotes cannot be called smooth if we want.
But in the relative comparison of the test problem with the text (where for all values of the parameter from 0 to 40 only 1 value leads to a change of F, and it is unknown which of 49 parameters causes the change of F) the quotes are as smooth as a maiden's knee.
...I think I can find an hour over the weekend to bolt the algorithm to the problem.
The algorithm should be bolted on exactly as in the task. The same connection interface will be on the championship FF.
In other words, everyone who will be able to connect to the FF from the task and work with it, will be able to participate in the championship.