Let a and b be positive integers such that
n = 4 ∑ 2 2 0 0 − 1 ⌊ lo g 2 n ⌋ = a ⋅ 2 b ,
where a is an odd integer and b is a positive integer. What is a + b ?
This problem is posed by Pebrudal Z .
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.
A very clever way to add up this sum! Great job!
Oh, 2 k + 1 − 2 k = 2 k . I could have used that, but I was too much occupied by the pattern in which terms got cancelled. :)
First, we note that ⌊ lo g k ( n ) ⌋ = x where x is the largest integer power of k that is less than n . This is because k x < n < k x + 1 → x < lo g k ( n ) < x + 1 → ⌊ lo g k ( n ) ⌋ = x .
We split our desired sum in the following way: n = 2 2 ∑ 2 3 − 1 ⌊ lo g 2 ( n ) ⌋ + n = 2 3 ∑ 2 4 − 1 ⌊ lo g 2 ( n ) ⌋ + . . . + n = 2 1 9 9 ∑ 2 2 0 0 − 1 ⌊ lo g 2 ( n ) ⌋ . In the k th sum in this expression (beginning with k = 2 ), there are 2 k values being summed, and each of these values is k , as we showed above, and so our desired sum is k = 2 ∑ 1 9 9 k ⋅ 2 k .
The first few partial sums are 8 = ( 2 − 1 ) ⋅ 2 2 + 1 , 3 2 = ( 3 − 1 ) ⋅ 2 3 + 1 , 9 6 = ( 4 − 1 ) ⋅ 2 4 + 1 , . . . , which suggests that the k th partial sum is ( k − 1 ) ⋅ 2 k + 1 . We can prove this using induction.
We already have the base case, so we assume it's true for k = j and then show it's true for k = j + 1 :
n = 2 ∑ j + 1 n ⋅ 2 n = n = 2 ∑ j n ⋅ 2 n + ( j + 1 ) ⋅ 2 j + 1 =
( j − 1 ) ⋅ 2 j + 1 + ( j + 1 ) ⋅ 2 j + 1 = ( 2 ⋅ j ) ⋅ 2 j + 1 =
( ( j + 1 ) − 1 ) ⋅ 2 ( j + 1 ) + 1
QED .
Thus, our desired sum is ( 1 9 9 − 1 ) ⋅ 2 1 9 9 + 1 = 9 9 ⋅ 2 2 0 1 , and so the answer is 9 9 + 2 0 1 = 3 0 0 .
Yes, this works. Nice job!
Log in to reply
I'm really worst at log questions. can i get some learning techniques from Brilliant?
Log in to reply
If you want to practice using the basic properties of logarithms, check out our practice section . You can find much simpler problems than this one, that allow you to practice using logs here .
I think this isn't really a log question; if you rewrite the sum to use the variable k = ⌊ lo g 2 n ⌋ then the log goes away. Also, you can often change a log into an exponent by raising both sides as a power of the base of the log.
n = 4 ∑ 2 2 0 0 − 1 ⌊ lo g 2 n ⌋
= 2 + 2 + … + 2 + 3 + 3 + … + 3 + … + 1 9 9 + 1 9 9 + … + 1 9 9
= 2 × 2 2 + 3 × 2 3 + 4 × 2 4 + … + 1 9 9 × 2 1 9 9
= k = 2 ∑ 1 9 9 k 2 k
Since i = 0 ∑ n − 1 i a i = ( 1 − a ) 2 a − n a n + ( n − 1 ) a n + 1 then k = 2 ∑ n − 1 k 2 k = ( k − 2 ) 2 k
Hence,
k = 2 ∑ 1 9 9 k 2 k = ( 2 0 0 − 2 ) × 2 2 0 0 = 9 9 × 2 2 0 1
a = 9 9 , b = 2 0 1 , a + b = 3 0 0
To evaluate that sum by direct algebra without using the given identity:
N N + 2 N = k = 2 ∑ k = 1 9 9 2 k k = k = 1 ∑ k = 1 9 9 2 k k = i = 1 ∑ i = 1 9 9 j = i ∑ j = 1 9 9 2 j = i = 1 ∑ i = 1 9 9 ( j = 0 ∑ j = 1 9 9 2 j − j = 0 ∑ j = i − 1 2 j ) = i = 1 ∑ i = 1 9 9 ( ( 2 2 0 0 − 1 ) − ( 2 i − 1 ) ) = i = 1 ∑ i = 1 9 9 ( 2 2 0 0 − 2 i ) = 1 9 9 ⋅ 2 2 0 0 − i = 1 ∑ i = 1 9 9 2 i = 1 9 9 ⋅ 2 2 0 0 − ( 2 2 0 0 − 2 ) = 1 9 8 ⋅ 2 2 0 0 = 9 9 ⋅ 2 2 0 1
Good job!
For the i = 0 ∑ n − 1 i a i , are you using the summation of an arithmetico-geometric sequence formula, or some other trick?
Log in to reply
I would use i = 0 ∑ n − 1 i a i = a i = 0 ∑ n − 1 d a d a i = a d a d i = 0 ∑ n − 1 a i = a d a d a − 1 a n − 1 which leads to the RHS of that(requires a bit of calculus, that is the quotient rule). Anyway I did something similar using generating functions.
I think that is the generalized usage of AGP sequence formula.
Yes, I did the summation with arithmetic-geometric formula
n = 4 ∑ 2 2 0 0 − 1 ⌊ l o g 2 n ⌋
= n = 2 2 ∑ 2 3 − 1 2 + n = 2 3 ∑ 2 4 − 1 3 + n = 2 4 ∑ 2 5 − 1 4 + ⋯ + n = 2 1 9 8 ∑ 2 1 9 9 − 1 1 9 8 + n = 2 1 9 9 ∑ 2 2 0 0 − 1 1 9 9
= 2 ∗ 2 2 + 3 ∗ 2 3 + 4 ∗ 2 4 + ⋯ + 1 9 8 ∗ 2 1 9 8 + 1 9 9 ∗ 2 1 9 9
Now, this is a simple Arithmetico - Geometrico Sequence , whose sum comes out to be 1 9 8 ∗ 2 2 0 0 = 9 9 ∗ 2 2 0 1 . Hence, a + b = 3 0 0
⌊ l o g 2 n ⌋ = k, for n ∈ { 2 k , 2 k + 1 , . . . , 2 k + 1 − 1 }
This set contain 2 k elements.
Then, the sum S = ∑ n = 4 2 2 0 0 − 1 ⌊ l o g 2 n ⌋ = ∑ k = 2 1 9 9 k × 2 k
S = 2 ∑ k = 2 1 9 9 k × 2 k − 1
S = lim a → 2 2 ∑ k = 2 1 9 9 k × a k − 1
S = lim a → 2 2 ∑ k = 2 1 9 9 d a d ( a k )
S = lim a → 2 d a d ( 2 ∑ k = 2 1 9 9 k × a k )
S = lim a → 2 d a d 2 a − 1 a 2 0 0 − a 2
S = 9 9 × 2 2 0 1
Therefore, a = 9 9 and b = 2 0 1 → a + b = 3 0 0 .
Check your equations, particularly the fourth line from the bottom.
Sorry! Don't have 'k' in the fourth line from the bottom. Thks, Calvin L.
Problem Loading...
Note Loading...
Set Loading...
Observe that for n = 2 k to ( 2 k + 1 − 1 ) (i.e., for ( 2 k + 1 − 2 k ) number of terms), ⌊ lo g 2 n ⌋ = k for some positive integer k . [This is a trivial observation]
(Note that ⌊ lo g 2 ( 2 2 0 0 − 1 ) ⌋ = 1 9 9 )
So we may transform the summation as
k = 2 ∑ 1 9 9 k ∗ ( 2 k + 1 − 2 k )
= 2 ( 2 3 − 2 2 ) + 3 ( 2 4 − 2 3 ) + 4 ( 2 5 − 2 4 ) + . . . + 1 9 9 ( 2 2 0 0 − 2 1 9 9 )
= 1 9 9 ⋅ 2 2 0 0 − 2 2 − ( 2 2 + 2 3 + 2 4 + . . . + 2 1 9 9 )
= 1 9 9 ⋅ 2 2 0 0 − 2 2 − ( 2 2 0 0 − 2 2 ) = 1 9 8 ⋅ 2 2 0 0 = 9 9 ⋅ 2 2 0 1 .
So a + b is 300 .