Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

There is a theorem In Terence Tao's Additive Combinatorics:

Proposition 2.2 (Exact inverse sum set theorem) Suppose that $A$, $B$ are additive sets with common ambient group Z . Then the following are equivalent:

  1. $|A + B| = |A|$;
  2. there exists a finite subgroup $G$ of $Z$ such that $B$ is contained in a coset of $G$, and A is a union of cosets of $G$.

It is straight forward to show that 1 implies 2. But how can we get the reverse direction work?

share|cite|improve this question
Does this $Z$ equal to the group $\mathbb Z$ of integers? (Well, that doesn't have any nontrivial finite subgroup..) And, just to be sure if I understand well the question, so $A$ and $B$ are subsets of $Z$, both closed under addition, right? – Berci Sep 30 '12 at 13:32
In the book, $Z$ is any abelian group, and $A$, $B$ are subsets not necessarily closed under addition. – only Sep 30 '12 at 15:11
aha, then how is additive sets meant? – Berci Sep 30 '12 at 15:18
up vote 0 down vote accepted

To prove 2 implies 1: Let $B \subset b+G$ and $A = \cup (a_i+G).$ Then $B+A$ will be $\cup (a_i+b+G)$, which has the same magnitude.

To prove 1 implies 2: Clearly we can translate $B$ without affecting the truth of $|A+B| = |B|$, so WLOG $B$ contains 0. Let $G$ be the subgroup generated by $B.$ Now, $B$ is clearly contained in $G$ (and when we undo the translate this becomes $B$ is contained in a coset of $G$.) Now, note that $A+B$ contains $A$ (As $B$ contains 0.) Since $|A+B| = |A|$, this implies $A+B = A.$

Therefore, if $a \in A$ and $b \in B$, $a+b = a'$ for some $a' \in A.$ Repeating this, we get that for all $a \in A$ and $b$ a linear combination of elements in $B$, $a+b \in A,$ or equivalent, for all $a \in A$ and $b \in G$, $a+b \in A.$

This tells you that $A = \displaystyle\cup_{a\in A} (a+G)$, which proves that $A$ is a union of cosets, as desired.

share|cite|improve this answer
Aha, thanks! I see what confuses me now. As you wrote, 1 implies that A+B = A, but somehow I was try to derive A+B=A directly from 2. But actually 1 is only about cardinality. – ablmf Sep 30 '12 at 16:12

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.