Bhaswar B. Bhattacharya, Anirban Chatterjee, Svante Janson
{"title":"Fluctuations of subgraph counts in graphon based random graphs","authors":"Bhaswar B. Bhattacharya, Anirban Chatterjee, Svante Janson","doi":"10.1017/s0963548322000335","DOIUrl":null,"url":null,"abstract":"<p>Given a graphon <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline1.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$W$\n</span></span>\n</span>\n</span> and a finite simple graph <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline2.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span>, with vertex set <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline3.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$V(H)$\n</span></span>\n</span>\n</span>, denote by <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline4.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$X_n(H, W)$\n</span></span>\n</span>\n</span> the number of copies of <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline5.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span> in a <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline6.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$W$\n</span></span>\n</span>\n</span>-random graph on <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline7.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$n$\n</span></span>\n</span>\n</span> vertices. The asymptotic distribution of <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline8.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$X_n(H, W)$\n</span></span>\n</span>\n</span> was recently obtained by Hladký, Pelekis, and Šileikis [17] in the case where <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline9.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span> is a clique. In this paper, we extend this result to any fixed graph <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline10.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span>. Towards this we introduce a notion of <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline11.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span>-regularity of graphons and show that if the graphon <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline12.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$W$\n</span></span>\n</span>\n</span> is not <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline13.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span>-regular, then <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline14.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$X_n(H, W)$\n</span></span>\n</span>\n</span> has Gaussian fluctuations with scaling <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline15.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$n^{|V(H)|-\\frac{1}{2}}$\n</span></span>\n</span>\n</span>. On the other hand, if <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline16.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$W$\n</span></span>\n</span>\n</span> is <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline17.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span>-regular, then the fluctuations are of order <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline18.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$n^{|V(H)|-1}$\n</span></span>\n</span>\n</span> and the limiting distribution of <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline19.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$X_n(H, W)$\n</span></span>\n</span>\n</span> can have both Gaussian and non-Gaussian components, where the non-Gaussian component is a (possibly) infinite weighted sum of centred chi-squared random variables with the weights determined by the spectral properties of a graphon derived from <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline20.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$W$\n</span></span>\n</span>\n</span>. Our proofs use the asymptotic theory of generalised <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline21.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$U$\n</span></span>\n</span>\n</span>-statistics developed by Janson and Nowicki [22]. We also investigate the structure of <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline22.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span>-regular graphons for which either the Gaussian or the non-Gaussian component of the limiting distribution (but not both) is degenerate. Interestingly, there are also <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline23.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H$\n</span></span>\n</span>\n</span>-regular graphons <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline24.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$W$\n</span></span>\n</span>\n</span> for which both the Gaussian or the non-Gaussian components are degenerate, that is, <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline25.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$X_n(H, W)$\n</span></span>\n</span>\n</span> has a degenerate limit even under the scaling <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline26.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$n^{|V(H)|-1}$\n</span></span>\n</span>\n</span>. We give an example of this degeneracy with <span>\n<span>\n<img data-mimesubtype=\"png\" data-type=\"\" src=\"https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20230406132554462-0824:S0963548322000335:S0963548322000335_inline27.png\"/>\n<span data-mathjax-type=\"texmath\"><span>\n$H=K_{1, 3}$\n</span></span>\n</span>\n</span> (the 3-star) and also establish non-degeneracy in a few examples. This naturally leads to interesting open questions on higher order degeneracies.</p>","PeriodicalId":10503,"journal":{"name":"Combinatorics, Probability and Computing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Combinatorics, Probability and Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1017/s0963548322000335","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Given a graphon
$W$
and a finite simple graph
$H$
, with vertex set
$V(H)$
, denote by
$X_n(H, W)$
the number of copies of
$H$
in a
$W$
-random graph on
$n$
vertices. The asymptotic distribution of
$X_n(H, W)$
was recently obtained by Hladký, Pelekis, and Šileikis [17] in the case where
$H$
is a clique. In this paper, we extend this result to any fixed graph
$H$
. Towards this we introduce a notion of
$H$
-regularity of graphons and show that if the graphon
$W$
is not
$H$
-regular, then
$X_n(H, W)$
has Gaussian fluctuations with scaling
$n^{|V(H)|-\frac{1}{2}}$
. On the other hand, if
$W$
is
$H$
-regular, then the fluctuations are of order
$n^{|V(H)|-1}$
and the limiting distribution of
$X_n(H, W)$
can have both Gaussian and non-Gaussian components, where the non-Gaussian component is a (possibly) infinite weighted sum of centred chi-squared random variables with the weights determined by the spectral properties of a graphon derived from
$W$
. Our proofs use the asymptotic theory of generalised
$U$
-statistics developed by Janson and Nowicki [22]. We also investigate the structure of
$H$
-regular graphons for which either the Gaussian or the non-Gaussian component of the limiting distribution (but not both) is degenerate. Interestingly, there are also
$H$
-regular graphons
$W$
for which both the Gaussian or the non-Gaussian components are degenerate, that is,
$X_n(H, W)$
has a degenerate limit even under the scaling
$n^{|V(H)|-1}$
. We give an example of this degeneracy with
$H=K_{1, 3}$
(the 3-star) and also establish non-degeneracy in a few examples. This naturally leads to interesting open questions on higher order degeneracies.