A Note on the Frankl Conjecture
Maysam Maysami Sadr

TL;DR
This paper explores variants of the Frankl conjecture, a well-known unsolved problem in combinatorics, by introducing and partially justifying new conjectural forms.
Contribution
It proposes and provides initial justification for several variants of the Frankl conjecture, expanding the scope of potential approaches.
Findings
Introduction of new variants of the Frankl conjecture
Partial theoretical justification for these variants
Potential implications for future research in combinatorics
Abstract
The Frankl conjecture (called also union-closed sets conjecture) is one of the famous unsolved conjectures in combinatorics of finite sets. In this short note, we introduce and to some extent justify some variants of the Frankl conjecture.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGraph theory and applications · Spectral Theory in Mathematical Physics
A Note on the Frankl Conjecture
Maysam Maysami Sadr Email: [email protected] Department of Mathematics, Institute for Advanced Studies in Basic Sciences, Zanjan, Iran
Abstract
The Frankl conjecture (called also union-closed sets conjecture) is one of the famous unsolved conjectures in combinatorics of finite sets. It has been formulated since mid-1970s as follows: If is a union-closed family of subsets of a finite set , which contains a nonempty subset of , then there exists such that belongs to at least half of the members of . In this short note, we introduce and to some extent justify some variants of the Frankl conjecture.
MSC 2010. Primary: 05D05. Secondary: 05A20.
Keywords. Union-closed sets conjecture, extremal set theory, combinatorics of finite sets, binomial expansion, combinatorics inequality.
Let be a finite nonempty set and let denote the powerset of . By a union-closed family on we mean a set such that , and such that if then . Frankl’s conjecture (FC for short), that is also called union-closed sets conjecture, is as follows.
Conjecture 1**.**
FC:* If is a union-closed family on a finite nonempty set , then there exists such that belongs to at least half of the members of .*
For history of FC, some basic related results, and some equivalent formulations of the conjecture in terms of graphs and lattices, we refer the reader to [1, 2] and references therein. In this short note, we introduce some natural variants of FC.
Let us begin by some notations and definitions. For a set , we denote by the cardinal of . If is a union-closed family on , we denote by the symbol . The class of all union-closed families on finite nonempty sets is denoted by .
Definition 2**.**
Let and be two natural numbers with . A family is called -separated if and for every distinct elements , there exists such that and . The class of all -separated union-closed families is denoted by .
By definition, any union-closed family is -separated; thus . Also, is the class of all union-closed families with . As a trivial example, for every finite set with , any union-closed family on , containing the union-closed family , is -separated.
For every two natural numbers and as above, let be the supremum of all positive real numbers satisfying the following condition: For every , there exists a set with such that there are at least members of satisfying . If there is no such an , then is defined to be [math]. Note that the ‘supremum’ in the definition of is actually a ‘maximum’.
Theorem 3**.**
Let be natural numbers such that and . The following four inequalities are satisfied.
- (1)
** 2. (2)
** 3. (3)
** 4. (4)
**
Proof.
(1) Let be a finite set with . Then, . It is easily checked that for every subset with , the number of those members of with , is equal to . This shows that (1) is satisfied.
(2) Let . Then, it is easily checked that also belongs to . Thus, there exists with , such that if the family is defined by
[TABLE]
then . Let the real number be such that . Suppose that the family is defined to be the empty family if , and otherwise,
[TABLE]
Also, let
[TABLE]
It is clear that coincides with the disjoint union . Thus,
[TABLE]
On the other hand, it is easily checked that is a union-closed family on the set and . Thus, there exists with such that for at least members of , . Let be the number of those members of with the property . Then,
[TABLE]
The above inequality together with show that (2) is satisfied.
(3) Let . Then, , and there exists a set with such that is contained in at least members of . Let
[TABLE]
Therefore, we have . On the other hand, it is easily seen that . Thus, there exists a set with such that there are at least members of satisfying . Therefore, has elements and there exists at least members of satisfying . This implies that (3) is satisfied.
(4) follows directly from (3). ∎
Now, we introduce our variant of the Frankl conjecture.
Conjecture 4**.**
FC of order :* If then there exists a subset with such that the number of those members of with , is at least . In other words, we have . In particular, , , and .*
It is clear that FC of order , is the original Frankl conjecture. We have suggested Conjecture 4, from inequality (1) in Theorem 3, and special cases of the conjecture of orders and .
Theorem 5**.**
Let be arbitrary natural numbers. The following statements are satisfied.
- (1)
Validity of FC of orders and imply validity of FC of order . In particular, validity of the Frankl conjecture implies validity of FC of order . 2. (2)
Validity of FC of orders and imply validity of FC of order . In particular, validity of the Frankl conjecture implies validity of FC of order .
Proof.
(1) follows from inequalities (1) and (2) of Theorem 3. (2) follows from inequalities (1) and (4) of Theorem 3. ∎
It is easily seen that the following two inequalities are satisfied:
[TABLE]
[TABLE]
Hence, Conjecture 4 is compatible with inequalities (2) and (3) of Theorem 3. This is another justification for Conjecture 4.
We suggest that FC of order is also satisfied in a bigger class of union-closed families than : Let be a union-closed family such that . We say that is weakly -separated if the following condition is satisfied: For any distinct elements , there exists such that and . we denote the class of all weakly -separated union-closed families by . It is clear that . We let the real number be exactly defined as except that this time the supremum being taken over . Obviously, we have .
Conjecture 6**.**
Strong FC of order :* .*
Note that the statements and proofs of Theorems 3 and 5 hold if the symbols are replaced by , and the phrase ‘FC of order’ by ‘strong FC of order’. Since (resp. ), strong FC and FC of order (resp. ) coincide.
Remark 7**.**
It is remarked that we have checked the validity of Conjectures 4 and 6 for families with .
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] H. Bruhn, O. Schaudt, The journey of the union-closed sets conjecture , Graphs Combin. 31, no. 6 (2015): 2043–2074. (ar Xiv:1309.3297 [math.CO])
- 2[2] B. Poonen, Union-closed families , Journal of Combinatorial Theory, Series A 59, no. 2 (1992): 253–268.
