n → ∞ lim ( 1 4 1 / 1 4 + 2 4 1 / 1 4 + 1 / 2 4 + … + n 4 1 / 1 4 + 1 / 2 4 + 1 / 3 4 + … + 1 / n 4 )
Given the limit above is equal to b a π c for positive integers a , b , c with a and b coprime, find the value of a + b + c .
Details and Assumptions :
You are given that k = 1 ∑ ∞ k 2 n 1 = ( − 1 ) n + 1 2 ( 2 n ) ! B 2 n ⋅ ( 2 π ) 2 n where B n is the Bernoulli numbers.
You may use the following values: B 0 = 1 , B 2 = 6 1 , B 4 = − 3 0 1 , B 6 = 4 2 1 , B 8 = − 3 0 1 .
Bonus : Generalize this.
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
For the bonus it can be generalised to:
i = 1 ∑ ∞ ( i a 1 k = 1 ∑ i k a 1 ) = 2 ζ ( a ) 2 + ζ ( 2 a )
Log in to reply
Well, that just becomes so obvious, I think after this. But yeah, a nice generalization.
Log in to reply
Yes it does. I just thought I'd point it out anyway.
Also with: 2 1 n = 1 ∑ ∞ n 4 τ ( n ) − 2 1 n = 1 ∑ ∞ ( n 2 ) 4 1
Did you mean: 2 1 n = 1 ∑ ∞ n 4 τ ( n ) + 2 1 n = 1 ∑ ∞ ( n 2 ) 4 1 ?
Log in to reply
@Isaac Buckley – Thanks for noting that typo. Edited!
Oh haha, I didn't know you have posted a solution! Thumbs up +1
that is easy
Problem Loading...
Note Loading...
Set Loading...
So, the problem is actually -
i = 1 ∑ ∞ ( i 4 1 k = 1 ∑ i k 4 1 )
Well, that looks easy now, right? Well, yeah.
We are going to transform the sum such that it becomes easy for us to compute.
n = 1 ∑ ∞ n 4 ⌊ 2 τ ( n ) ⌋
where τ ( n ) is the divisor function .
You may see how that might have come. It came out, however, because we are multiplying two numbers whatsoever to get a new number and then summing all. So, the number of divisors a number has and the number of times it will come out in the sum would be just half. Also, that is not the case with squares where number of divisors is odd.
So, after we have got why and how that is so, it is easy to remove that floor(since we(more specifically I) don't know how to compute directly from the floor function).
2 1 n = 1 ∑ ∞ n 4 τ ( n ) + 2 1 n = 1 ∑ ∞ ( n 2 ) 4 1
The first sum can be written as n = 1 ∑ ∞ n 4 1 ( p = 1 ∑ ∞ p 4 1 ) = ζ ( 4 ) 2
And the second of course becomes ζ ( 8 ) .
So, our total sum becomes - 2 1 ( ζ ( 4 ) 2 + ζ ( 8 ) )
= 1 1 3 4 0 0 1 3 π 8