Pure maths, physics, logic (braingames.ru): non-trade-related brain games - page 172
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
The sum is calculated on the fly. The obsolete value is removed (which has gone beyond the averaging period) and the newly arrived value is added. "Sliding amount".
This is how all of our muvings are calculated. However, Rashid has somehow dug out a faster algorithm for calculating moving average. We didn't use it because we cannot strictly prove the adequacy - the values seem to be the same, but we don't know how this algorithm may behave in the future
Thank you. "Got it." It all makes perfect sense. :)
Can we have a look at the algorithm found by Rashid?
I join in the request. As I have little idea what can be faster than a one-pass algorithm - two additions and one division at each iteration step.
For head-to-head algorithms O(N^2) and above, there are indeed sometimes non-obviously fast counterparts.
For example, used in an intermediate FFT step to compute millions of Pearson correlation coefficient values in less than a second.
Well, yeah, it's kind of complicated. But I haven't got it right yet (haven't looked at it):
It's not complicated, there's a simple solution:
A mini-brain enters and is equally likely to take the jersey of the eliminated MM or the last one. If that doesn't happen he takes the jersey of any other one. When it is the turn of the MM whose jersey he has taken, he can also with equal probability take the jersey of the first (Rejected) or the last. Taking the jersey of the first one causes the latter to take his jersey, and taking the jersey of the latter respectively deprives him of this opportunity. Each time, the one who enters and does not find his T-shirt with equal probability can take the first or the last one, the others are of no interest to us.
The real challenge
A bear fell into a trap hole 19.617 metres deep. Its time of fall was 2 seconds. What colour was the bear?
А. White (polar bear)
B. Brown
C. Black
D. Black and brown (Malay bear)
E. Grey (grizzly bear)
There are a dozen parameters, which I want to find with the optimizer. In order not to use a billion parameters, I search by function (let's say parabola), i.e. I approximate and instead of a billion parameters - now there are only 4-5 - they are shift on x-axis, y, scale on x and y, and order. But now a question has arisen - how to specify ranges for these four five parameters in the optimiser? how to calculate this?
Who wants half a grand in 15 minutes?
http://habrahabr.ru/post/193308/?utm_source=twitterfeed&utm_medium=habrahabr&utm_campaign=twitter
http://sphotos-a.ak.fbcdn.net/hphotos-ak-prn2/q79/s720x720/1384223_10153412503770533_1695215820_n.jpg
I haven't solved that problem. I've already solved it.
Translation from English:
This problem can be solved in 5-10 minutes by a preschooler, in an hour by a coder, and in a man with higher education... Well, try it yourself!
http://sphotos-a.ak.fbcdn.net/hphotos-ak-prn2/q79/s720x720/1384223_10153412503770533_1695215820_n.jpg
I haven't solved that problem. I've already solved it.
Translation from English:
This problem can be solved in 5-10 minutes by a preschooler, in an hour by a coder, and in a man with higher education... in short, try it yourself!
2581 = 2
// 6 minutes, I'll pass for pre-schooler. ;)
It doesn't say whether it's right or not.
And I'm sure a 'pre-schooler' solves the problem even faster. But that's just wrong.
Realistically, this problem is very good to illustrate "tweaking to the curve", which happens in robot optimization.
You can put ANY number in the answer. For any number there is a pattern. I found for 0,1,2 - bored further, but I'm sure the other 7 digits will do as well.