Set of sets such that any three of them has nonempty intersection. Show all of them have nonempty...











up vote
2
down vote

favorite
2













Given $2^{n-1}$ subsets of a set with $n$ elements with the property that any three have nonempty intersection, prove that the intersection of all the sets is nonempty.




My attempt:
Let $|X|=n$ and $S subset mathcal{P}(X)$ be the above mentioned set. If the intersection of all the sets in $S$ is nonempty and contains W.L.O.G $a in X$, then the set of all subsets of $X$ containing $a$ satisfy the requirement.
enter image description here










share|cite|improve this question















closed as off-topic by José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici Nov 15 at 10:53


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici

If this question can be reworded to fit the rules in the help center, please edit the question.













  • Interesting question? Where did you see this question, and what have you tried so far?
    – Frpzzd
    Nov 14 at 23:12










  • your attempt assumes the result? I think it is a great question and might have a great answer. (see below). what is the motivation?
    – user136920
    Nov 15 at 11:07

















up vote
2
down vote

favorite
2













Given $2^{n-1}$ subsets of a set with $n$ elements with the property that any three have nonempty intersection, prove that the intersection of all the sets is nonempty.




My attempt:
Let $|X|=n$ and $S subset mathcal{P}(X)$ be the above mentioned set. If the intersection of all the sets in $S$ is nonempty and contains W.L.O.G $a in X$, then the set of all subsets of $X$ containing $a$ satisfy the requirement.
enter image description here










share|cite|improve this question















closed as off-topic by José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici Nov 15 at 10:53


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici

If this question can be reworded to fit the rules in the help center, please edit the question.













  • Interesting question? Where did you see this question, and what have you tried so far?
    – Frpzzd
    Nov 14 at 23:12










  • your attempt assumes the result? I think it is a great question and might have a great answer. (see below). what is the motivation?
    – user136920
    Nov 15 at 11:07















up vote
2
down vote

favorite
2









up vote
2
down vote

favorite
2






2






Given $2^{n-1}$ subsets of a set with $n$ elements with the property that any three have nonempty intersection, prove that the intersection of all the sets is nonempty.




My attempt:
Let $|X|=n$ and $S subset mathcal{P}(X)$ be the above mentioned set. If the intersection of all the sets in $S$ is nonempty and contains W.L.O.G $a in X$, then the set of all subsets of $X$ containing $a$ satisfy the requirement.
enter image description here










share|cite|improve this question
















Given $2^{n-1}$ subsets of a set with $n$ elements with the property that any three have nonempty intersection, prove that the intersection of all the sets is nonempty.




My attempt:
Let $|X|=n$ and $S subset mathcal{P}(X)$ be the above mentioned set. If the intersection of all the sets in $S$ is nonempty and contains W.L.O.G $a in X$, then the set of all subsets of $X$ containing $a$ satisfy the requirement.
enter image description here







combinatorics elementary-set-theory contest-math problem-solving






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Nov 15 at 0:02









Batominovski

31.6k23188




31.6k23188










asked Nov 14 at 23:09









mathnoob

73211




73211




closed as off-topic by José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici Nov 15 at 10:53


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici

If this question can be reworded to fit the rules in the help center, please edit the question.




closed as off-topic by José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici Nov 15 at 10:53


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – José Carlos Santos, B. Mehta, max_zorn, ArsenBerk, Claude Leibovici

If this question can be reworded to fit the rules in the help center, please edit the question.












  • Interesting question? Where did you see this question, and what have you tried so far?
    – Frpzzd
    Nov 14 at 23:12










  • your attempt assumes the result? I think it is a great question and might have a great answer. (see below). what is the motivation?
    – user136920
    Nov 15 at 11:07




















  • Interesting question? Where did you see this question, and what have you tried so far?
    – Frpzzd
    Nov 14 at 23:12










  • your attempt assumes the result? I think it is a great question and might have a great answer. (see below). what is the motivation?
    – user136920
    Nov 15 at 11:07


















Interesting question? Where did you see this question, and what have you tried so far?
– Frpzzd
Nov 14 at 23:12




Interesting question? Where did you see this question, and what have you tried so far?
– Frpzzd
Nov 14 at 23:12












your attempt assumes the result? I think it is a great question and might have a great answer. (see below). what is the motivation?
– user136920
Nov 15 at 11:07






your attempt assumes the result? I think it is a great question and might have a great answer. (see below). what is the motivation?
– user136920
Nov 15 at 11:07












1 Answer
1






active

oldest

votes

















up vote
2
down vote



accepted










The claim clearly holds for $n=1$ and $n=2$. (See Hagen von Eitzen's comment below. With a different interpretation of the problem statement, the claim is false for $n=1$ and $n=2$.) In what follows, we assume that $ngeq 3$.



Let $S$ be a set with $n$ elements and $mathcal{F}$ a family of subsets of $S$ with the required properties. That is, $mathcal{F}$ satisfies the condition that $|mathcal{F}|=2^{n-1}$ and that any three (not necessarily distinct) sets in $mathcal{F}$ have a nonempty intersection.



Consider the vector space $V:=mathbb{F}_2^S$ with basis $big{e_s,|,sin Sbig}$ over the field $mathbb{F}_2$ with $2$ elements. We identify each subset $Xsubseteq S$ with the vector $$e_X:=sum_{sin X},e_s,.$$
In particular, $e_{{s}}=e_s$ for all $sin S$, and $e_emptyset=0$. Equip $V$ with the nondegenerate inner product $langle_,_rangle$ defined by
$$leftlangle e_X,e_Yrightrangle :=|Xcap Y|$$
for every $X,Ysubseteq S$ (here, $|Xcap Y|$ is considered modulo $2$).



We claim that $W:=mathbb{F}_2^Ssetminus big{e_X,big|,Xinmathcal{F}big}$ is a vector subspace of $V$. That is, $W$ is closed under addition and scalar multiplication. Clearly, $W$ is closed under scalar multiplication since $0=e_emptysetin W$ and $mathbb{F}_2={0,1}$. We are left with closure under addition.



For a subset $X$ of $S$, write $X^complement$ for $Ssetminus X$. It can be easily seen that, for each $Xsubseteq S$, either $Xinmathcal{F}$ or $X^complement in mathcal{F}$, but not both.



Suppose $A$ and $B$ are subsets of $S$ such that $e_Ain W$ and $e_Bin W$. Then, $A^complementinmathcal{F}$ and $B^complement inmathcal{F}$. Now, $e_A+e_B=e_{Atriangle B}$, where $triangle$ is the symmetric difference. Because $$(Atriangle B)cap A^complement cap B^complement=emptyset,,$$ $Atriangle B$ cannot be in $mathcal{F}$. Thus, $e_{Atriangle B}in W$.



Consequently, $W$ is indeed a subspace of $V$ with dimension $n-1$, i.e., with codimension $1$. There exists a unique linear functional $varphi:Vto mathbb{F}_2$ with kernel $W$. Hence, $varphi=langle e_T,_rangle$ for some nonempty $Tsubseteq S$ (this is because $langle_,_rangle$ is a nondegenerate symmetric bilinear form on $V$). It remains to show that $T$ is a singleton.



If $T$ contains at least three pairwise distinct elements, say $a$, $b$, and $c$, then take $A:={a}$, $B:={b}$, and $C:={a,b,c}$. Since $e_A$, $e_B$, and $e_C$ are not in $ker(varphi)$, $A$, $B$, and $C$ must be in $mathcal{F}$, but $Acap Bcap C=emptyset$. This is a contradiction, so $|T|leq 2$.



If $|T|=2$, say, $T={a,b}$, then we take $cin Ssetminus T$ (noting that $|S|=ngeq 3>|T|$), and set $A:={a}$, $B:={b}$, and $C:={a,c}$ and apply the same argument as before to arrive at a contradiction. Ergo, $T$ is a singleton: $T={t}$ for some $tin S$. This proves that every set in $mathcal{F}$ contains $t$.






share|cite|improve this answer























  • First thank you for the amazing answer!! So $W$ has codimention 1 because $V/W cong W^{bot}$ has only two vectors(hence of dimention 1) and $ varphi (x)=langle e_T, _ rangle$ because any linear functional can be represented as inner product with some vector?
    – mathnoob
    Nov 15 at 1:44












  • It's the other way around. Since $dim_{mathbb{F}_2}(W)=n-1=dim_{mathbb{F}_2}(V)-1$, $W$ as codiension $1$, and therefore, $V/W$ is a vector space over $mathbb{F}_2$ with only two elements (and so it is isomorphic to $mathbb{F}_2$). For the second question, yes, given that the inner product is nondegenerate.
    – Batominovski
    Nov 15 at 6:46








  • 1




    Arguably, the claim is clearly false for $n=1$ and $n=2$ - if one interpretes the condition as "the intersection of any distinct three sets ..."
    – Hagen von Eitzen
    Nov 15 at 7:01






  • 1




    @HagenvonEitzen That is why I added "not necessarily distinct" in my answer to eliminate these cases.
    – Batominovski
    Nov 15 at 7:06








  • 1




    @Batominovski Agreed - the potential ambiguity is rather in the problem statement than in your answer (or: one must be aware that "distinct" is not to be assumed unless explicitly stated)
    – Hagen von Eitzen
    Nov 15 at 7:10


















1 Answer
1






active

oldest

votes








1 Answer
1






active

oldest

votes









active

oldest

votes






active

oldest

votes








up vote
2
down vote



accepted










The claim clearly holds for $n=1$ and $n=2$. (See Hagen von Eitzen's comment below. With a different interpretation of the problem statement, the claim is false for $n=1$ and $n=2$.) In what follows, we assume that $ngeq 3$.



Let $S$ be a set with $n$ elements and $mathcal{F}$ a family of subsets of $S$ with the required properties. That is, $mathcal{F}$ satisfies the condition that $|mathcal{F}|=2^{n-1}$ and that any three (not necessarily distinct) sets in $mathcal{F}$ have a nonempty intersection.



Consider the vector space $V:=mathbb{F}_2^S$ with basis $big{e_s,|,sin Sbig}$ over the field $mathbb{F}_2$ with $2$ elements. We identify each subset $Xsubseteq S$ with the vector $$e_X:=sum_{sin X},e_s,.$$
In particular, $e_{{s}}=e_s$ for all $sin S$, and $e_emptyset=0$. Equip $V$ with the nondegenerate inner product $langle_,_rangle$ defined by
$$leftlangle e_X,e_Yrightrangle :=|Xcap Y|$$
for every $X,Ysubseteq S$ (here, $|Xcap Y|$ is considered modulo $2$).



We claim that $W:=mathbb{F}_2^Ssetminus big{e_X,big|,Xinmathcal{F}big}$ is a vector subspace of $V$. That is, $W$ is closed under addition and scalar multiplication. Clearly, $W$ is closed under scalar multiplication since $0=e_emptysetin W$ and $mathbb{F}_2={0,1}$. We are left with closure under addition.



For a subset $X$ of $S$, write $X^complement$ for $Ssetminus X$. It can be easily seen that, for each $Xsubseteq S$, either $Xinmathcal{F}$ or $X^complement in mathcal{F}$, but not both.



Suppose $A$ and $B$ are subsets of $S$ such that $e_Ain W$ and $e_Bin W$. Then, $A^complementinmathcal{F}$ and $B^complement inmathcal{F}$. Now, $e_A+e_B=e_{Atriangle B}$, where $triangle$ is the symmetric difference. Because $$(Atriangle B)cap A^complement cap B^complement=emptyset,,$$ $Atriangle B$ cannot be in $mathcal{F}$. Thus, $e_{Atriangle B}in W$.



Consequently, $W$ is indeed a subspace of $V$ with dimension $n-1$, i.e., with codimension $1$. There exists a unique linear functional $varphi:Vto mathbb{F}_2$ with kernel $W$. Hence, $varphi=langle e_T,_rangle$ for some nonempty $Tsubseteq S$ (this is because $langle_,_rangle$ is a nondegenerate symmetric bilinear form on $V$). It remains to show that $T$ is a singleton.



If $T$ contains at least three pairwise distinct elements, say $a$, $b$, and $c$, then take $A:={a}$, $B:={b}$, and $C:={a,b,c}$. Since $e_A$, $e_B$, and $e_C$ are not in $ker(varphi)$, $A$, $B$, and $C$ must be in $mathcal{F}$, but $Acap Bcap C=emptyset$. This is a contradiction, so $|T|leq 2$.



If $|T|=2$, say, $T={a,b}$, then we take $cin Ssetminus T$ (noting that $|S|=ngeq 3>|T|$), and set $A:={a}$, $B:={b}$, and $C:={a,c}$ and apply the same argument as before to arrive at a contradiction. Ergo, $T$ is a singleton: $T={t}$ for some $tin S$. This proves that every set in $mathcal{F}$ contains $t$.






share|cite|improve this answer























  • First thank you for the amazing answer!! So $W$ has codimention 1 because $V/W cong W^{bot}$ has only two vectors(hence of dimention 1) and $ varphi (x)=langle e_T, _ rangle$ because any linear functional can be represented as inner product with some vector?
    – mathnoob
    Nov 15 at 1:44












  • It's the other way around. Since $dim_{mathbb{F}_2}(W)=n-1=dim_{mathbb{F}_2}(V)-1$, $W$ as codiension $1$, and therefore, $V/W$ is a vector space over $mathbb{F}_2$ with only two elements (and so it is isomorphic to $mathbb{F}_2$). For the second question, yes, given that the inner product is nondegenerate.
    – Batominovski
    Nov 15 at 6:46








  • 1




    Arguably, the claim is clearly false for $n=1$ and $n=2$ - if one interpretes the condition as "the intersection of any distinct three sets ..."
    – Hagen von Eitzen
    Nov 15 at 7:01






  • 1




    @HagenvonEitzen That is why I added "not necessarily distinct" in my answer to eliminate these cases.
    – Batominovski
    Nov 15 at 7:06








  • 1




    @Batominovski Agreed - the potential ambiguity is rather in the problem statement than in your answer (or: one must be aware that "distinct" is not to be assumed unless explicitly stated)
    – Hagen von Eitzen
    Nov 15 at 7:10















up vote
2
down vote



accepted










The claim clearly holds for $n=1$ and $n=2$. (See Hagen von Eitzen's comment below. With a different interpretation of the problem statement, the claim is false for $n=1$ and $n=2$.) In what follows, we assume that $ngeq 3$.



Let $S$ be a set with $n$ elements and $mathcal{F}$ a family of subsets of $S$ with the required properties. That is, $mathcal{F}$ satisfies the condition that $|mathcal{F}|=2^{n-1}$ and that any three (not necessarily distinct) sets in $mathcal{F}$ have a nonempty intersection.



Consider the vector space $V:=mathbb{F}_2^S$ with basis $big{e_s,|,sin Sbig}$ over the field $mathbb{F}_2$ with $2$ elements. We identify each subset $Xsubseteq S$ with the vector $$e_X:=sum_{sin X},e_s,.$$
In particular, $e_{{s}}=e_s$ for all $sin S$, and $e_emptyset=0$. Equip $V$ with the nondegenerate inner product $langle_,_rangle$ defined by
$$leftlangle e_X,e_Yrightrangle :=|Xcap Y|$$
for every $X,Ysubseteq S$ (here, $|Xcap Y|$ is considered modulo $2$).



We claim that $W:=mathbb{F}_2^Ssetminus big{e_X,big|,Xinmathcal{F}big}$ is a vector subspace of $V$. That is, $W$ is closed under addition and scalar multiplication. Clearly, $W$ is closed under scalar multiplication since $0=e_emptysetin W$ and $mathbb{F}_2={0,1}$. We are left with closure under addition.



For a subset $X$ of $S$, write $X^complement$ for $Ssetminus X$. It can be easily seen that, for each $Xsubseteq S$, either $Xinmathcal{F}$ or $X^complement in mathcal{F}$, but not both.



Suppose $A$ and $B$ are subsets of $S$ such that $e_Ain W$ and $e_Bin W$. Then, $A^complementinmathcal{F}$ and $B^complement inmathcal{F}$. Now, $e_A+e_B=e_{Atriangle B}$, where $triangle$ is the symmetric difference. Because $$(Atriangle B)cap A^complement cap B^complement=emptyset,,$$ $Atriangle B$ cannot be in $mathcal{F}$. Thus, $e_{Atriangle B}in W$.



Consequently, $W$ is indeed a subspace of $V$ with dimension $n-1$, i.e., with codimension $1$. There exists a unique linear functional $varphi:Vto mathbb{F}_2$ with kernel $W$. Hence, $varphi=langle e_T,_rangle$ for some nonempty $Tsubseteq S$ (this is because $langle_,_rangle$ is a nondegenerate symmetric bilinear form on $V$). It remains to show that $T$ is a singleton.



If $T$ contains at least three pairwise distinct elements, say $a$, $b$, and $c$, then take $A:={a}$, $B:={b}$, and $C:={a,b,c}$. Since $e_A$, $e_B$, and $e_C$ are not in $ker(varphi)$, $A$, $B$, and $C$ must be in $mathcal{F}$, but $Acap Bcap C=emptyset$. This is a contradiction, so $|T|leq 2$.



If $|T|=2$, say, $T={a,b}$, then we take $cin Ssetminus T$ (noting that $|S|=ngeq 3>|T|$), and set $A:={a}$, $B:={b}$, and $C:={a,c}$ and apply the same argument as before to arrive at a contradiction. Ergo, $T$ is a singleton: $T={t}$ for some $tin S$. This proves that every set in $mathcal{F}$ contains $t$.






share|cite|improve this answer























  • First thank you for the amazing answer!! So $W$ has codimention 1 because $V/W cong W^{bot}$ has only two vectors(hence of dimention 1) and $ varphi (x)=langle e_T, _ rangle$ because any linear functional can be represented as inner product with some vector?
    – mathnoob
    Nov 15 at 1:44












  • It's the other way around. Since $dim_{mathbb{F}_2}(W)=n-1=dim_{mathbb{F}_2}(V)-1$, $W$ as codiension $1$, and therefore, $V/W$ is a vector space over $mathbb{F}_2$ with only two elements (and so it is isomorphic to $mathbb{F}_2$). For the second question, yes, given that the inner product is nondegenerate.
    – Batominovski
    Nov 15 at 6:46








  • 1




    Arguably, the claim is clearly false for $n=1$ and $n=2$ - if one interpretes the condition as "the intersection of any distinct three sets ..."
    – Hagen von Eitzen
    Nov 15 at 7:01






  • 1




    @HagenvonEitzen That is why I added "not necessarily distinct" in my answer to eliminate these cases.
    – Batominovski
    Nov 15 at 7:06








  • 1




    @Batominovski Agreed - the potential ambiguity is rather in the problem statement than in your answer (or: one must be aware that "distinct" is not to be assumed unless explicitly stated)
    – Hagen von Eitzen
    Nov 15 at 7:10













up vote
2
down vote



accepted







up vote
2
down vote



accepted






The claim clearly holds for $n=1$ and $n=2$. (See Hagen von Eitzen's comment below. With a different interpretation of the problem statement, the claim is false for $n=1$ and $n=2$.) In what follows, we assume that $ngeq 3$.



Let $S$ be a set with $n$ elements and $mathcal{F}$ a family of subsets of $S$ with the required properties. That is, $mathcal{F}$ satisfies the condition that $|mathcal{F}|=2^{n-1}$ and that any three (not necessarily distinct) sets in $mathcal{F}$ have a nonempty intersection.



Consider the vector space $V:=mathbb{F}_2^S$ with basis $big{e_s,|,sin Sbig}$ over the field $mathbb{F}_2$ with $2$ elements. We identify each subset $Xsubseteq S$ with the vector $$e_X:=sum_{sin X},e_s,.$$
In particular, $e_{{s}}=e_s$ for all $sin S$, and $e_emptyset=0$. Equip $V$ with the nondegenerate inner product $langle_,_rangle$ defined by
$$leftlangle e_X,e_Yrightrangle :=|Xcap Y|$$
for every $X,Ysubseteq S$ (here, $|Xcap Y|$ is considered modulo $2$).



We claim that $W:=mathbb{F}_2^Ssetminus big{e_X,big|,Xinmathcal{F}big}$ is a vector subspace of $V$. That is, $W$ is closed under addition and scalar multiplication. Clearly, $W$ is closed under scalar multiplication since $0=e_emptysetin W$ and $mathbb{F}_2={0,1}$. We are left with closure under addition.



For a subset $X$ of $S$, write $X^complement$ for $Ssetminus X$. It can be easily seen that, for each $Xsubseteq S$, either $Xinmathcal{F}$ or $X^complement in mathcal{F}$, but not both.



Suppose $A$ and $B$ are subsets of $S$ such that $e_Ain W$ and $e_Bin W$. Then, $A^complementinmathcal{F}$ and $B^complement inmathcal{F}$. Now, $e_A+e_B=e_{Atriangle B}$, where $triangle$ is the symmetric difference. Because $$(Atriangle B)cap A^complement cap B^complement=emptyset,,$$ $Atriangle B$ cannot be in $mathcal{F}$. Thus, $e_{Atriangle B}in W$.



Consequently, $W$ is indeed a subspace of $V$ with dimension $n-1$, i.e., with codimension $1$. There exists a unique linear functional $varphi:Vto mathbb{F}_2$ with kernel $W$. Hence, $varphi=langle e_T,_rangle$ for some nonempty $Tsubseteq S$ (this is because $langle_,_rangle$ is a nondegenerate symmetric bilinear form on $V$). It remains to show that $T$ is a singleton.



If $T$ contains at least three pairwise distinct elements, say $a$, $b$, and $c$, then take $A:={a}$, $B:={b}$, and $C:={a,b,c}$. Since $e_A$, $e_B$, and $e_C$ are not in $ker(varphi)$, $A$, $B$, and $C$ must be in $mathcal{F}$, but $Acap Bcap C=emptyset$. This is a contradiction, so $|T|leq 2$.



If $|T|=2$, say, $T={a,b}$, then we take $cin Ssetminus T$ (noting that $|S|=ngeq 3>|T|$), and set $A:={a}$, $B:={b}$, and $C:={a,c}$ and apply the same argument as before to arrive at a contradiction. Ergo, $T$ is a singleton: $T={t}$ for some $tin S$. This proves that every set in $mathcal{F}$ contains $t$.






share|cite|improve this answer














The claim clearly holds for $n=1$ and $n=2$. (See Hagen von Eitzen's comment below. With a different interpretation of the problem statement, the claim is false for $n=1$ and $n=2$.) In what follows, we assume that $ngeq 3$.



Let $S$ be a set with $n$ elements and $mathcal{F}$ a family of subsets of $S$ with the required properties. That is, $mathcal{F}$ satisfies the condition that $|mathcal{F}|=2^{n-1}$ and that any three (not necessarily distinct) sets in $mathcal{F}$ have a nonempty intersection.



Consider the vector space $V:=mathbb{F}_2^S$ with basis $big{e_s,|,sin Sbig}$ over the field $mathbb{F}_2$ with $2$ elements. We identify each subset $Xsubseteq S$ with the vector $$e_X:=sum_{sin X},e_s,.$$
In particular, $e_{{s}}=e_s$ for all $sin S$, and $e_emptyset=0$. Equip $V$ with the nondegenerate inner product $langle_,_rangle$ defined by
$$leftlangle e_X,e_Yrightrangle :=|Xcap Y|$$
for every $X,Ysubseteq S$ (here, $|Xcap Y|$ is considered modulo $2$).



We claim that $W:=mathbb{F}_2^Ssetminus big{e_X,big|,Xinmathcal{F}big}$ is a vector subspace of $V$. That is, $W$ is closed under addition and scalar multiplication. Clearly, $W$ is closed under scalar multiplication since $0=e_emptysetin W$ and $mathbb{F}_2={0,1}$. We are left with closure under addition.



For a subset $X$ of $S$, write $X^complement$ for $Ssetminus X$. It can be easily seen that, for each $Xsubseteq S$, either $Xinmathcal{F}$ or $X^complement in mathcal{F}$, but not both.



Suppose $A$ and $B$ are subsets of $S$ such that $e_Ain W$ and $e_Bin W$. Then, $A^complementinmathcal{F}$ and $B^complement inmathcal{F}$. Now, $e_A+e_B=e_{Atriangle B}$, where $triangle$ is the symmetric difference. Because $$(Atriangle B)cap A^complement cap B^complement=emptyset,,$$ $Atriangle B$ cannot be in $mathcal{F}$. Thus, $e_{Atriangle B}in W$.



Consequently, $W$ is indeed a subspace of $V$ with dimension $n-1$, i.e., with codimension $1$. There exists a unique linear functional $varphi:Vto mathbb{F}_2$ with kernel $W$. Hence, $varphi=langle e_T,_rangle$ for some nonempty $Tsubseteq S$ (this is because $langle_,_rangle$ is a nondegenerate symmetric bilinear form on $V$). It remains to show that $T$ is a singleton.



If $T$ contains at least three pairwise distinct elements, say $a$, $b$, and $c$, then take $A:={a}$, $B:={b}$, and $C:={a,b,c}$. Since $e_A$, $e_B$, and $e_C$ are not in $ker(varphi)$, $A$, $B$, and $C$ must be in $mathcal{F}$, but $Acap Bcap C=emptyset$. This is a contradiction, so $|T|leq 2$.



If $|T|=2$, say, $T={a,b}$, then we take $cin Ssetminus T$ (noting that $|S|=ngeq 3>|T|$), and set $A:={a}$, $B:={b}$, and $C:={a,c}$ and apply the same argument as before to arrive at a contradiction. Ergo, $T$ is a singleton: $T={t}$ for some $tin S$. This proves that every set in $mathcal{F}$ contains $t$.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Nov 15 at 7:11

























answered Nov 14 at 23:57









Batominovski

31.6k23188




31.6k23188












  • First thank you for the amazing answer!! So $W$ has codimention 1 because $V/W cong W^{bot}$ has only two vectors(hence of dimention 1) and $ varphi (x)=langle e_T, _ rangle$ because any linear functional can be represented as inner product with some vector?
    – mathnoob
    Nov 15 at 1:44












  • It's the other way around. Since $dim_{mathbb{F}_2}(W)=n-1=dim_{mathbb{F}_2}(V)-1$, $W$ as codiension $1$, and therefore, $V/W$ is a vector space over $mathbb{F}_2$ with only two elements (and so it is isomorphic to $mathbb{F}_2$). For the second question, yes, given that the inner product is nondegenerate.
    – Batominovski
    Nov 15 at 6:46








  • 1




    Arguably, the claim is clearly false for $n=1$ and $n=2$ - if one interpretes the condition as "the intersection of any distinct three sets ..."
    – Hagen von Eitzen
    Nov 15 at 7:01






  • 1




    @HagenvonEitzen That is why I added "not necessarily distinct" in my answer to eliminate these cases.
    – Batominovski
    Nov 15 at 7:06








  • 1




    @Batominovski Agreed - the potential ambiguity is rather in the problem statement than in your answer (or: one must be aware that "distinct" is not to be assumed unless explicitly stated)
    – Hagen von Eitzen
    Nov 15 at 7:10


















  • First thank you for the amazing answer!! So $W$ has codimention 1 because $V/W cong W^{bot}$ has only two vectors(hence of dimention 1) and $ varphi (x)=langle e_T, _ rangle$ because any linear functional can be represented as inner product with some vector?
    – mathnoob
    Nov 15 at 1:44












  • It's the other way around. Since $dim_{mathbb{F}_2}(W)=n-1=dim_{mathbb{F}_2}(V)-1$, $W$ as codiension $1$, and therefore, $V/W$ is a vector space over $mathbb{F}_2$ with only two elements (and so it is isomorphic to $mathbb{F}_2$). For the second question, yes, given that the inner product is nondegenerate.
    – Batominovski
    Nov 15 at 6:46








  • 1




    Arguably, the claim is clearly false for $n=1$ and $n=2$ - if one interpretes the condition as "the intersection of any distinct three sets ..."
    – Hagen von Eitzen
    Nov 15 at 7:01






  • 1




    @HagenvonEitzen That is why I added "not necessarily distinct" in my answer to eliminate these cases.
    – Batominovski
    Nov 15 at 7:06








  • 1




    @Batominovski Agreed - the potential ambiguity is rather in the problem statement than in your answer (or: one must be aware that "distinct" is not to be assumed unless explicitly stated)
    – Hagen von Eitzen
    Nov 15 at 7:10
















First thank you for the amazing answer!! So $W$ has codimention 1 because $V/W cong W^{bot}$ has only two vectors(hence of dimention 1) and $ varphi (x)=langle e_T, _ rangle$ because any linear functional can be represented as inner product with some vector?
– mathnoob
Nov 15 at 1:44






First thank you for the amazing answer!! So $W$ has codimention 1 because $V/W cong W^{bot}$ has only two vectors(hence of dimention 1) and $ varphi (x)=langle e_T, _ rangle$ because any linear functional can be represented as inner product with some vector?
– mathnoob
Nov 15 at 1:44














It's the other way around. Since $dim_{mathbb{F}_2}(W)=n-1=dim_{mathbb{F}_2}(V)-1$, $W$ as codiension $1$, and therefore, $V/W$ is a vector space over $mathbb{F}_2$ with only two elements (and so it is isomorphic to $mathbb{F}_2$). For the second question, yes, given that the inner product is nondegenerate.
– Batominovski
Nov 15 at 6:46






It's the other way around. Since $dim_{mathbb{F}_2}(W)=n-1=dim_{mathbb{F}_2}(V)-1$, $W$ as codiension $1$, and therefore, $V/W$ is a vector space over $mathbb{F}_2$ with only two elements (and so it is isomorphic to $mathbb{F}_2$). For the second question, yes, given that the inner product is nondegenerate.
– Batominovski
Nov 15 at 6:46






1




1




Arguably, the claim is clearly false for $n=1$ and $n=2$ - if one interpretes the condition as "the intersection of any distinct three sets ..."
– Hagen von Eitzen
Nov 15 at 7:01




Arguably, the claim is clearly false for $n=1$ and $n=2$ - if one interpretes the condition as "the intersection of any distinct three sets ..."
– Hagen von Eitzen
Nov 15 at 7:01




1




1




@HagenvonEitzen That is why I added "not necessarily distinct" in my answer to eliminate these cases.
– Batominovski
Nov 15 at 7:06






@HagenvonEitzen That is why I added "not necessarily distinct" in my answer to eliminate these cases.
– Batominovski
Nov 15 at 7:06






1




1




@Batominovski Agreed - the potential ambiguity is rather in the problem statement than in your answer (or: one must be aware that "distinct" is not to be assumed unless explicitly stated)
– Hagen von Eitzen
Nov 15 at 7:10




@Batominovski Agreed - the potential ambiguity is rather in the problem statement than in your answer (or: one must be aware that "distinct" is not to be assumed unless explicitly stated)
– Hagen von Eitzen
Nov 15 at 7:10



Popular posts from this blog

How do I know what Microsoft account the skydrive app is syncing to?

When does type information flow backwards in C++?

Grease: Live!