A is a non-empty proper subset of Z with at least two elements, and with the following property:
If a and b are distinct elements of A , then − a , a + 2 b , and a + 1 9 are also elements of A .
Find a = 0 a ∈ A min ( ∣ a ∣ ) .
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.
But simply speaking,. If a = 2 and b=1; then we have smallest element in A as 0
Log in to reply
In the problem statement, Othman Slassi wrote a = 0 . It is printed smallish underneath "min", but it is there!
Denote m = a ∈ A min ( ∣ a ∣ ) we first proof this claim:
Claim: A is a subgroup of ( Z , + )
Let a , b ∈ A we have : ( ∀ m ≥ 1 ) : a + 2 m b ∈ A ( Just use induction, easy!) so a + 2 0 b ∈ A . By easy induction we also have: ( ∀ x ∈ A ) ( ∀ n ∈ Z ) : x + 1 9 n ∈ A for n = − b and x = a + 2 0 b we find: a + b ∈ A □
A is a subgroup of ( Z , + ) so A = m Z witch means that if x ∈ A : m ∣ x but we have also m ∣ ( x + 1 9 ) since x + 1 9 ∈ A so m ∣ 1 9 . m = 1 9 , since we cannot have m = 1 because then we would have A = Z .
Why did you put in + 2 ^ m * b € A ...? You could simply put a + 2mb € A (by induction), and then x = a + 2 * 10b and n = -b and get the same prove without needing Fermat's Little Theoreme.
And there's a typo, I think you wanted to say x + 19n rather than a +19n.
Very nice prove.
Log in to reply
YES, thank you you're absolutly right!
Log in to reply
Can you make those edits directly to your solution? Thanks
Assume that there is an element b in A that is not divisible by 1 9 . Then 2 b is also not divisible by 1 9 , and by Fermat’s little theorem ( 2 b ) 1 8 ≡ 1 m o d 1 9 . Then through multiple uses of the rule a + 2 b we can start with an element a (not equal to b ) and arrive at an element c = a + ( 2 b ) 1 8 ≡ a + 1 m o d 1 9 , and through multiple uses of the rule − a and a + 1 9 we can obtain element d = a + 1 .
In other words, if b is not divisible by 1 9 , we can carry out the above steps on element a to obtain element a + 1 , and again on element a + 1 to obtain element a + 2 , and so on. Using this and the rule of − a , all the integers Z can be obtained in A . But this contradicts the fact that A is a proper subset of Z .
Therefore, all the elements in A must be divisible by 1 9 , and the smallest non-zero multiple of 1 9 is 1 9 .
Going all the way to a + ( 2 b ) 1 8 is overkill... There always exists 1 ≤ x ≤ 1 8 such that 2 b x ≡ 1 mod 19, unless b is a multiple of 19. (And if b is a multiple of 19, then a + ( 2 b ) 1 8 won't be equivalent to a + 1 mod 19, either.)
Log in to reply
Thanks for the tip! I did not know about that theorem, whereas I did know about Fermat's Little Theorem. I suppose I could improve my solution to replace "and by Fermat’s little theorem ( 2 b ) 1 8 ≡ 1 m o d 1 9 . Then through multiple uses of the rule a + 2 b we can start with an element a (not equal to b ) and arrive at an element c = a + ( 2 b ) 1 8 ≡ a + 1 m o d 1 9 " with "there always exists 1 ≤ x ≤ 1 8 such that 2 b x ≡ 1 m o d 1 9 . Then through multiple uses of the rule a + 2 b we can start with an element a (not equal to b ) and arrive at an element c = a + 2 b x ≡ a + 1 m o d 1 9 ", but then it would be very similar to your solution, and I was hoping to present something different. (Using Fermat's Little Theorem is also how I originally solved the problem.)
I did a slight variation on this: for a+2b, take b=a mod 19, i.e. if a is in the set then so is 3a ; since 3 and 19 are coprime this implies that if there is a non-zero (mod 19) set member then all integers are in the set.
Problem Loading...
Note Loading...
Set Loading...
Short proof
We have A = A + 1 9 Z and for any b ∈ A , A = A + 2 b Z . Thus A = A + 1 9 Z + 2 b Z = A + gcd ( 1 9 , 2 b ) Z , but since A = Z , we must have gcd ( 1 9 , 2 b ) > 1 so that 1 9 ∣ b and A ⊆ 1 9 Z . However, if A = ∅ then also A ⊇ b + 1 9 Z = 1 9 Z , so that in fact A = 1 9 Z and the desired minimum is 1 9 .
Longer version for those who don't like concise set notation.
(1) If a ∈ A , then a + 1 9 n ∈ A for all integers n . Proof:
For n = 0 it is trivial.
For n > 0 , by induction. If a ′ = a + 1 9 ( n − 1 ) ∈ A , then according to the rule, a ′ + 1 9 = a + 1 9 n ∈ A .
For n < 0 , by induction. If a ′ = a + 1 9 ( n + 1 ) ∈ A , then − a ′ ∈ A , so − a ′ + 1 9 ∈ A , so − ( − a ′ + 1 9 ) = a ′ − 1 9 = a + 1 9 n ∈ A .
(2) If a , b ∈ A , then a + 2 b n ∈ A for all integers n . Proof:
Assuming a = b , use a similar reasoning by induction as above.
The case a = b is easily avoided by observing that a + 1 9 ∈ A , so a + 1 9 + 2 b n ∈ A for all n , so a + 1 9 + 2 b n − 1 9 = a + 2 b n ∈ A for all n .
(3) Suppose b ∈ A is not a multiple of 19, and suppose a ∈ A . Let c ∈ Z . Then c ∈ A . Proof:
gcd ( 2 b , 1 9 ) = 1 .
From the (extended) Eucleidan algorithm, there exist integers n 1 , n 2 such that c − a = 2 b n 1 + 1 9 n 2 .
Since a ∈ A , we know from (2) that a + 2 b n 1 ∈ A , and from (1) that a + 2 b n 1 + 1 9 n 2 = c ∈ A .
(4) We are told that A = Z , so that there must be an integer c ∈ A . Using the converse of (3), we conclude that there does not exist b ∈ A that is not a multiple of 19. It follows that A only contains multiples of 19.
(5) Since A is not empty, it contains at least one such multiple of 19, say a = 1 9 m . But from (1) we know that then all integers a + 1 9 n = 1 9 ( n + m ) are elements of A . In other words, A is precisely the set of all multiples of 19 ( A = 1 9 Z ).
It is now easy to conclude that the answer to the problem is 1 9 .