{"title":"独立横轴的度数标准和稳定性","authors":"Penny Haxell, Ronen Wdowinski","doi":"10.1002/jgt.23085","DOIUrl":null,"url":null,"abstract":"<p>An <i>independent transversal</i> (IT) in a graph <span></span><math>\n <semantics>\n <mrow>\n <mi>G</mi>\n </mrow>\n <annotation> $G$</annotation>\n </semantics></math> with a given vertex partition <span></span><math>\n <semantics>\n <mrow>\n <mi>P</mi>\n </mrow>\n <annotation> ${\\mathscr{P}}$</annotation>\n </semantics></math> is an independent set of vertices of <span></span><math>\n <semantics>\n <mrow>\n <mi>G</mi>\n </mrow>\n <annotation> $G$</annotation>\n </semantics></math> (i.e., it induces no edges), that consists of one vertex from each part (<i>block</i>) of <span></span><math>\n <semantics>\n <mrow>\n <mi>P</mi>\n </mrow>\n <annotation> ${\\mathscr{P}}$</annotation>\n </semantics></math>. Over the years, various criteria have been established that guarantee the existence of an IT, often given in terms of <span></span><math>\n <semantics>\n <mrow>\n <mi>P</mi>\n </mrow>\n <annotation> ${\\mathscr{P}}$</annotation>\n </semantics></math> being <span></span><math>\n <semantics>\n <mrow>\n <mi>t</mi>\n </mrow>\n <annotation> $t$</annotation>\n </semantics></math>-<i>thick</i>, meaning all blocks have size at least <span></span><math>\n <semantics>\n <mrow>\n <mi>t</mi>\n </mrow>\n <annotation> $t$</annotation>\n </semantics></math>. One such result, obtained recently by Wanless and Wood, is based on the <i>maximum average block degree</i> <span></span><math>\n <semantics>\n <mrow>\n <mi>b</mi>\n <mrow>\n <mo>(</mo>\n <mrow>\n <mi>G</mi>\n <mo>,</mo>\n <mi>P</mi>\n </mrow>\n <mo>)</mo>\n </mrow>\n <mo>=</mo>\n <mi>max</mi>\n <mrow>\n <mo>{</mo>\n <mrow>\n <msub>\n <mo>∑</mo>\n <mrow>\n <mi>u</mi>\n <mo>∈</mo>\n <mi>U</mi>\n </mrow>\n </msub>\n <mi>d</mi>\n <mrow>\n <mo>(</mo>\n <mi>u</mi>\n <mo>)</mo>\n </mrow>\n <mo>∕</mo>\n <mo>∣</mo>\n <mi>U</mi>\n <mo>∣</mo>\n <mo>:</mo>\n <mi>U</mi>\n <mo>∈</mo>\n <mi>P</mi>\n </mrow>\n <mo>}</mo>\n </mrow>\n </mrow>\n <annotation> $b(G,{\\mathscr{P}})=\\max \\{{\\sum }_{u\\in U}d(u)\\unicode{x02215}| U| :U\\in {\\mathscr{P}}\\}$</annotation>\n </semantics></math>. They proved that if <span></span><math>\n <semantics>\n <mrow>\n <mi>b</mi>\n <mrow>\n <mo>(</mo>\n <mrow>\n <mi>G</mi>\n <mo>,</mo>\n <mi>P</mi>\n </mrow>\n <mo>)</mo>\n </mrow>\n <mo>≤</mo>\n <mi>t</mi>\n <mo>∕</mo>\n <mn>4</mn>\n </mrow>\n <annotation> $b(G,{\\mathscr{P}})\\le t\\unicode{x02215}4$</annotation>\n </semantics></math> then an IT exists. Resolving a problem posed by Groenland, Kaiser, Treffers and Wales (who showed that the ratio 1/4 is best possible), here we give a full characterization of pairs <span></span><math>\n <semantics>\n <mrow>\n <mrow>\n <mo>(</mo>\n <mrow>\n <mi>α</mi>\n <mo>,</mo>\n <mi>β</mi>\n </mrow>\n <mo>)</mo>\n </mrow>\n </mrow>\n <annotation> $(\\alpha ,\\beta )$</annotation>\n </semantics></math> such that the following holds for every <span></span><math>\n <semantics>\n <mrow>\n <mi>t</mi>\n <mo>></mo>\n <mn>0</mn>\n </mrow>\n <annotation> $t\\gt 0$</annotation>\n </semantics></math>: whenever <span></span><math>\n <semantics>\n <mrow>\n <mi>G</mi>\n </mrow>\n <annotation> $G$</annotation>\n </semantics></math> is a graph with maximum degree <span></span><math>\n <semantics>\n <mrow>\n <mi>Δ</mi>\n <mrow>\n <mo>(</mo>\n <mi>G</mi>\n <mo>)</mo>\n </mrow>\n <mo>≤</mo>\n <mi>α</mi>\n <mi>t</mi>\n </mrow>\n <annotation> ${\\rm{\\Delta }}(G)\\le \\alpha t$</annotation>\n </semantics></math>, and <span></span><math>\n <semantics>\n <mrow>\n <mi>P</mi>\n </mrow>\n <annotation> ${\\mathscr{P}}$</annotation>\n </semantics></math> is a <span></span><math>\n <semantics>\n <mrow>\n <mi>t</mi>\n </mrow>\n <annotation> $t$</annotation>\n </semantics></math>-thick vertex partition of <span></span><math>\n <semantics>\n <mrow>\n <mi>G</mi>\n </mrow>\n <annotation> $G$</annotation>\n </semantics></math> such that <span></span><math>\n <semantics>\n <mrow>\n <mi>b</mi>\n <mrow>\n <mo>(</mo>\n <mrow>\n <mi>G</mi>\n <mo>,</mo>\n <mi>P</mi>\n </mrow>\n <mo>)</mo>\n </mrow>\n <mo>≤</mo>\n <mi>β</mi>\n <mi>t</mi>\n </mrow>\n <annotation> $b(G,{\\mathscr{P}})\\le \\beta t$</annotation>\n </semantics></math>, there exists an IT of <span></span><math>\n <semantics>\n <mrow>\n <mi>G</mi>\n </mrow>\n <annotation> $G$</annotation>\n </semantics></math> with respect to <span></span><math>\n <semantics>\n <mrow>\n <mi>P</mi>\n </mrow>\n <annotation> ${\\mathscr{P}}$</annotation>\n </semantics></math>. Our proof makes use of another previously known criterion for the existence of ITs that involve the topological connectedness of the independence complex of graphs, and establishes a general technical theorem on the structure of graphs for which this parameter is bounded above by a known quantity. Our result interpolates between the criterion <span></span><math>\n <semantics>\n <mrow>\n <mi>b</mi>\n <mrow>\n <mo>(</mo>\n <mrow>\n <mi>G</mi>\n <mo>,</mo>\n <mi>P</mi>\n </mrow>\n <mo>)</mo>\n </mrow>\n <mo>≤</mo>\n <mi>t</mi>\n <mo>∕</mo>\n <mn>4</mn>\n </mrow>\n <annotation> $b(G,{\\mathscr{P}})\\le t\\unicode{x02215}4$</annotation>\n </semantics></math> and the old and frequently applied theorem that if <span></span><math>\n <semantics>\n <mrow>\n <mi>Δ</mi>\n <mrow>\n <mo>(</mo>\n <mi>G</mi>\n <mo>)</mo>\n </mrow>\n <mo>≤</mo>\n <mi>t</mi>\n <mo>∕</mo>\n <mn>2</mn>\n </mrow>\n <annotation> ${\\rm{\\Delta }}(G)\\le t\\unicode{x02215}2$</annotation>\n </semantics></math> then an IT exists. Using the same approach, we also extend a theorem of Aharoni, Holzman, Howard and Sprüssel, by giving a stability version of the latter result.</p>","PeriodicalId":16014,"journal":{"name":"Journal of Graph Theory","volume":"106 2","pages":"352-371"},"PeriodicalIF":0.9000,"publicationDate":"2024-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jgt.23085","citationCount":"0","resultStr":"{\"title\":\"Degree criteria and stability for independent transversals\",\"authors\":\"Penny Haxell, Ronen Wdowinski\",\"doi\":\"10.1002/jgt.23085\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>An <i>independent transversal</i> (IT) in a graph <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>G</mi>\\n </mrow>\\n <annotation> $G$</annotation>\\n </semantics></math> with a given vertex partition <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>P</mi>\\n </mrow>\\n <annotation> ${\\\\mathscr{P}}$</annotation>\\n </semantics></math> is an independent set of vertices of <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>G</mi>\\n </mrow>\\n <annotation> $G$</annotation>\\n </semantics></math> (i.e., it induces no edges), that consists of one vertex from each part (<i>block</i>) of <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>P</mi>\\n </mrow>\\n <annotation> ${\\\\mathscr{P}}$</annotation>\\n </semantics></math>. Over the years, various criteria have been established that guarantee the existence of an IT, often given in terms of <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>P</mi>\\n </mrow>\\n <annotation> ${\\\\mathscr{P}}$</annotation>\\n </semantics></math> being <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>t</mi>\\n </mrow>\\n <annotation> $t$</annotation>\\n </semantics></math>-<i>thick</i>, meaning all blocks have size at least <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>t</mi>\\n </mrow>\\n <annotation> $t$</annotation>\\n </semantics></math>. One such result, obtained recently by Wanless and Wood, is based on the <i>maximum average block degree</i> <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>b</mi>\\n <mrow>\\n <mo>(</mo>\\n <mrow>\\n <mi>G</mi>\\n <mo>,</mo>\\n <mi>P</mi>\\n </mrow>\\n <mo>)</mo>\\n </mrow>\\n <mo>=</mo>\\n <mi>max</mi>\\n <mrow>\\n <mo>{</mo>\\n <mrow>\\n <msub>\\n <mo>∑</mo>\\n <mrow>\\n <mi>u</mi>\\n <mo>∈</mo>\\n <mi>U</mi>\\n </mrow>\\n </msub>\\n <mi>d</mi>\\n <mrow>\\n <mo>(</mo>\\n <mi>u</mi>\\n <mo>)</mo>\\n </mrow>\\n <mo>∕</mo>\\n <mo>∣</mo>\\n <mi>U</mi>\\n <mo>∣</mo>\\n <mo>:</mo>\\n <mi>U</mi>\\n <mo>∈</mo>\\n <mi>P</mi>\\n </mrow>\\n <mo>}</mo>\\n </mrow>\\n </mrow>\\n <annotation> $b(G,{\\\\mathscr{P}})=\\\\max \\\\{{\\\\sum }_{u\\\\in U}d(u)\\\\unicode{x02215}| U| :U\\\\in {\\\\mathscr{P}}\\\\}$</annotation>\\n </semantics></math>. They proved that if <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>b</mi>\\n <mrow>\\n <mo>(</mo>\\n <mrow>\\n <mi>G</mi>\\n <mo>,</mo>\\n <mi>P</mi>\\n </mrow>\\n <mo>)</mo>\\n </mrow>\\n <mo>≤</mo>\\n <mi>t</mi>\\n <mo>∕</mo>\\n <mn>4</mn>\\n </mrow>\\n <annotation> $b(G,{\\\\mathscr{P}})\\\\le t\\\\unicode{x02215}4$</annotation>\\n </semantics></math> then an IT exists. Resolving a problem posed by Groenland, Kaiser, Treffers and Wales (who showed that the ratio 1/4 is best possible), here we give a full characterization of pairs <span></span><math>\\n <semantics>\\n <mrow>\\n <mrow>\\n <mo>(</mo>\\n <mrow>\\n <mi>α</mi>\\n <mo>,</mo>\\n <mi>β</mi>\\n </mrow>\\n <mo>)</mo>\\n </mrow>\\n </mrow>\\n <annotation> $(\\\\alpha ,\\\\beta )$</annotation>\\n </semantics></math> such that the following holds for every <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>t</mi>\\n <mo>></mo>\\n <mn>0</mn>\\n </mrow>\\n <annotation> $t\\\\gt 0$</annotation>\\n </semantics></math>: whenever <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>G</mi>\\n </mrow>\\n <annotation> $G$</annotation>\\n </semantics></math> is a graph with maximum degree <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>Δ</mi>\\n <mrow>\\n <mo>(</mo>\\n <mi>G</mi>\\n <mo>)</mo>\\n </mrow>\\n <mo>≤</mo>\\n <mi>α</mi>\\n <mi>t</mi>\\n </mrow>\\n <annotation> ${\\\\rm{\\\\Delta }}(G)\\\\le \\\\alpha t$</annotation>\\n </semantics></math>, and <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>P</mi>\\n </mrow>\\n <annotation> ${\\\\mathscr{P}}$</annotation>\\n </semantics></math> is a <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>t</mi>\\n </mrow>\\n <annotation> $t$</annotation>\\n </semantics></math>-thick vertex partition of <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>G</mi>\\n </mrow>\\n <annotation> $G$</annotation>\\n </semantics></math> such that <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>b</mi>\\n <mrow>\\n <mo>(</mo>\\n <mrow>\\n <mi>G</mi>\\n <mo>,</mo>\\n <mi>P</mi>\\n </mrow>\\n <mo>)</mo>\\n </mrow>\\n <mo>≤</mo>\\n <mi>β</mi>\\n <mi>t</mi>\\n </mrow>\\n <annotation> $b(G,{\\\\mathscr{P}})\\\\le \\\\beta t$</annotation>\\n </semantics></math>, there exists an IT of <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>G</mi>\\n </mrow>\\n <annotation> $G$</annotation>\\n </semantics></math> with respect to <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>P</mi>\\n </mrow>\\n <annotation> ${\\\\mathscr{P}}$</annotation>\\n </semantics></math>. Our proof makes use of another previously known criterion for the existence of ITs that involve the topological connectedness of the independence complex of graphs, and establishes a general technical theorem on the structure of graphs for which this parameter is bounded above by a known quantity. Our result interpolates between the criterion <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>b</mi>\\n <mrow>\\n <mo>(</mo>\\n <mrow>\\n <mi>G</mi>\\n <mo>,</mo>\\n <mi>P</mi>\\n </mrow>\\n <mo>)</mo>\\n </mrow>\\n <mo>≤</mo>\\n <mi>t</mi>\\n <mo>∕</mo>\\n <mn>4</mn>\\n </mrow>\\n <annotation> $b(G,{\\\\mathscr{P}})\\\\le t\\\\unicode{x02215}4$</annotation>\\n </semantics></math> and the old and frequently applied theorem that if <span></span><math>\\n <semantics>\\n <mrow>\\n <mi>Δ</mi>\\n <mrow>\\n <mo>(</mo>\\n <mi>G</mi>\\n <mo>)</mo>\\n </mrow>\\n <mo>≤</mo>\\n <mi>t</mi>\\n <mo>∕</mo>\\n <mn>2</mn>\\n </mrow>\\n <annotation> ${\\\\rm{\\\\Delta }}(G)\\\\le t\\\\unicode{x02215}2$</annotation>\\n </semantics></math> then an IT exists. Using the same approach, we also extend a theorem of Aharoni, Holzman, Howard and Sprüssel, by giving a stability version of the latter result.</p>\",\"PeriodicalId\":16014,\"journal\":{\"name\":\"Journal of Graph Theory\",\"volume\":\"106 2\",\"pages\":\"352-371\"},\"PeriodicalIF\":0.9000,\"publicationDate\":\"2024-02-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jgt.23085\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Graph Theory\",\"FirstCategoryId\":\"100\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1002/jgt.23085\",\"RegionNum\":3,\"RegionCategory\":\"数学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MATHEMATICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Graph Theory","FirstCategoryId":"100","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/jgt.23085","RegionNum":3,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MATHEMATICS","Score":null,"Total":0}
引用次数: 0
摘要
具有给定顶点分区 P ${mathscr{P}}$ 的图 G $G$ 中的独立横向(IT)是 G $G$ 的一个独立顶点集合(即它不引起任何边),它由 P ${\mathscr{P}}$ 的每个部分(块)的一个顶点组成。多年来,人们建立了各种标准来保证 IT 的存在,这些标准通常以 P ${mathscr{P}}$ 厚度为 t $t$ 的条件给出,即所有块的大小至少为 t $t$。其中一个结果是 Wanless 和 Wood 最近得到的,它基于最大平均块度 b ( G , P ) = max { ∑ u∈ U d ( u ) ∕ ∣ U ∣ : U∈ P }。 $b(G,{\mathscr{P}})=\max \{\{sum }_{u\in U}d(u)\unicode{x02215}| U| :U\in {\mathscr{P}}\}$ 。他们证明了如果 b ( G , P ) ≤ t ∕ 4 $b(G,{\mathscr{P}})\le t\unicode{x02215}4$ 则存在一个 IT。
Degree criteria and stability for independent transversals
An independent transversal (IT) in a graph with a given vertex partition is an independent set of vertices of (i.e., it induces no edges), that consists of one vertex from each part (block) of . Over the years, various criteria have been established that guarantee the existence of an IT, often given in terms of being -thick, meaning all blocks have size at least . One such result, obtained recently by Wanless and Wood, is based on the maximum average block degree . They proved that if then an IT exists. Resolving a problem posed by Groenland, Kaiser, Treffers and Wales (who showed that the ratio 1/4 is best possible), here we give a full characterization of pairs such that the following holds for every : whenever is a graph with maximum degree , and is a -thick vertex partition of such that , there exists an IT of with respect to . Our proof makes use of another previously known criterion for the existence of ITs that involve the topological connectedness of the independence complex of graphs, and establishes a general technical theorem on the structure of graphs for which this parameter is bounded above by a known quantity. Our result interpolates between the criterion and the old and frequently applied theorem that if then an IT exists. Using the same approach, we also extend a theorem of Aharoni, Holzman, Howard and Sprüssel, by giving a stability version of the latter result.
期刊介绍:
The Journal of Graph Theory is devoted to a variety of topics in graph theory, such as structural results about graphs, graph algorithms with theoretical emphasis, and discrete optimization on graphs. The scope of the journal also includes related areas in combinatorics and the interaction of graph theory with other mathematical sciences.
A subscription to the Journal of Graph Theory includes a subscription to the Journal of Combinatorial Designs .