Over all positive integers n and for all n numbers a 1 , a 2 , . . . , a n satisfying − 1 ≤ a 1 , a 2 , . . . , a n ≤ 1 and i = 1 ∑ n a i 3 = 0 , find the maximum value of n ∑ i = 1 n a i to 3 decimal places.
Inspired by this person
.
Here is an
easier version
.
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.
This approach is known as the tangent line inequality , which is useful in cases where the variables are separable (no cross terms).
The motivation for the inequality in the first line is consider the tangent to
y
=
x
3
, that passes through
(
−
1
,
−
1
)
. The point of tangent occurs at
x
=
2
1
. We then take
x
3
−
"equation for this tangent line" to get us a bound involving
x
3
and
x
.
Since we know that this is a cubic which passes through
x
=
−
1
, and is tangential at
x
=
2
1
, we can conclude that the relevant inequality is
(
x
−
2
1
)
2
(
x
−
(
−
1
)
)
≥
0
.
Great solution! Thanks!
how about if n is not divisible by 9?
Log in to reply
If so, you can not pick non-negative integers a , b satisfying a + b = n and b − 8 a = 0 , thus the maximum in such cases can't be 1 / 3 .
Log in to reply
What Steven meant is that this approach will not work.
I suspect there isn't an easy way of finding the maximum then (though we know it is strictly smaller then 1/3)
We want to maximize f ( x ) = j = 1 ∑ n sin x j subject to the constraint g ( x ) = j = 1 ∑ n sin 3 x j = 0 For simplicity, suppose that n ≥ 9 . Then the case sin x 1 = − 1 , sin x 2 = sin x 3 = ⋯ = sin x 9 = 2 1 , with sin x j = 0 for j > 9 , shows that the maximum possible value of f subject to this constraint will be positive.
Since there is no boundary to the domain for x , the maximum of f subject to the constraint g = 0 will be identified by the method of Lagrange multipliers, so there exists λ such that ∇ f − λ ∇ g = 0 , and hence cos x j ( 1 − 3 λ sin 2 x j ) = 0 1 ≤ j ≤ n If cos x j = 0 for all j , then sin x j = ± 1 for all j , which would imply that f = 0 . Thus, for a positive value of f , there must be a number 0 < μ < 1 such that sin x j ∈ { 1 , − 1 , μ , − μ } for all j . Thus there must exist nonnegative integers a , b , c , d such that a + b + c + d a − b + μ 3 ( c − d ) a − b + μ ( c − d ) = n = 0 = f (so that a of the sin x j are equal to 1 , b are equal to − 1 , c are equal to μ and d are equal to − μ ). Since f > 0 we deduce that c > d , and hence that b > a . Moreover b − a < c − d , and we have μ = ( c − d b − a ) 3 1 f = ( b − a ) 3 1 ( c − d ) 3 2 − ( b − a ) For a given choice of a , b , it is clear that f will be maximized by maximizing c − d , so we must have c = n − a − b and d = 0 , and hence f = ( b − a ) 3 1 ( n − a − b ) 3 2 − ( b − a ) Putting t = b − a , we see that 0 ≤ t ≤ n − 2 a and that f = t 3 1 ( n − 2 a − t ) 3 2 − t From this it is clear that, for a given choice of t , f is maximized by minimizing a , and so a = 0 . Thus we are picking a = d = 0 and b + c = n where b < c , and so we want to maximize f = b 3 1 ( n − b ) 3 2 − b 1 ≤ b < 2 1 n The function F ( u ) = u 3 1 ( 1 − u ) 3 2 − u 0 < u < 1 has its maximum value of 3 1 at u = 9 1 , and is increasing for u < 9 1 and decreasing for u > 9 1 . Thus we deduce that f will be equal to either F ( ⌊ 9 1 n ⌋ ) or else F ( ⌊ 9 1 n ⌋ + 1 ) , whichever is larger.
Thus the maximum value of n − 1 ∑ j = 1 n y j subject to the constraint ∑ j = 1 n y j 3 = 0 and ∣ y j ∣ ≤ 1 for all 1 ≤ j ≤ n is n 1 [ b 3 1 ( n − b ) 3 2 − b ] where b is either ⌊ 9 1 n ⌋ or ⌊ 9 1 n ⌋ + 1 , and this value is achieved by choosing b of the y j equal to − 1 , with the remaining n − b values of y j equal to ( n − b b ) 3 1 . This maximum value is less than or equal to 3 1 , being equal to 3 1 when 9 divides n . I have not investigated which of the two possible values of b need to be chosen for any other particular value of n , but in many cases we choose b to the integer closest to 9 1 n .
@Calvin Lin
This is a constrained optimization problem, with a Lagrangian of L ( a 1 , a 2 , ⋯ , a n ) = a 1 + a 2 ⋯ + a n − λ ⋅ ( a 1 3 + a 2 3 + ⋯ + a n 3 ) . Setting ∇ L = 0 yields a 1 = ± a 2 = ± a 3 = ⋯ ± a n . It's clearly unoptimal for that to be true, since f = n a 1 + a 2 + … a n = 0 .
Hence at least one of the values is on the boundary, namely a n = ± 1 . Considering the new function with its new constraints, we again have a 1 = ± a 2 = ⋯ = ± a n − 1 . Since having a k = − a j here makes no impact on the final sum but enlarges n , they must either all be equal or another one has to be on the boundary. Continuing in this fashion, we have x of the n numbers on the (same) boundary, and the rest equal. Since differentiating between the − 1 and 1 cases for the boundary simply changes f 's sign, we assume WLOG a 1 = a 2 = ⋯ = a x = − 1 .
Thus a x + 1 = a x + 2 = ⋯ = a n = ( n − x x ) 3 1 , and so f = n ( n − x ) 3 2 ⋅ x 3 1 − x . Letting x 3 1 = a and ( n − x ) 3 1 = k ⋅ a with k ≥ 1 , which is what our boundary condition a p ≤ 1 implies, we have f = a 3 ⋅ ( k 3 + 1 ) k 2 ⋅ a 3 − a 3 = k 2 − k + 1 k − 1 . This reaches its maximum for k = 2 , hence n − x = k 3 ⋅ x = 8 ⋅ x , making n = 9 ⋅ x and f = k 2 − k + 1 k − 1 = 3 1 , which is the answer.
Is there a non-Lagrangian solution for this problem?
Good solution anyways!
Log in to reply
Probably, seeing as I only really use the Lagrangian to show that the a 's must either be on the boundary or equal to each other. I think an argument can be made about there being an "optimal" value for the a 's not on the boundary, hence why they're all equal. Either way, I think using the Lagrangian is the easiest way to justify the problem rigorously.
The maximum of f ( x ) = n 1 [ x 3 1 ( n − x ) 3 2 − x ] over 0 ≤ x ≤ 2 1 n is indeed 3 1 , achieved when x = 9 1 n .
We need a slight nuance to your proof. If n is not a multiple of 9 , then the above maximum of f is not achieved at an integer value of x , and so the maximum value of n 1 ∑ j = 1 n a j , subject to the given constraints, is less than 3 1 . For example, when n = 1 0 the maximum is 1 0 1 ( 3 3 4 − 1 ) . However, the maximum value is exactly 3 1 whenever n is a multiple of 9 , and so (since the question also asks us to maximize f over n ) the answer of 3 1 is correct.
Problem Loading...
Note Loading...
Set Loading...
@Jon Haussmann 's proof of the special case generalises easily. Note that 4 a j 3 − 3 a j + 1 = ( 2 a j − 1 ) 2 ( a j + 1 ) ≥ 0 for all j , so that 3 j = 1 ∑ n a j ≤ 4 j = 1 ∑ n a j 3 + n = n and hence n 1 j = 1 ∑ n a j ≤ 3 1 This inequality will become an equality if all the a j can be chosen to be equal to either − 1 or 2 1 , which means that there exist non-negative integers a , b such that a + b = n and − a + 8 1 b = 0 . This is possible if n is divisible by 9 , with a = 9 1 n and b = 9 8 n . Thus, maximising over all n , see that the upper bound is 3 1 .