<?xml-model href='http://www.tei-c.org/release/xml/tei/custom/schema/relaxng/tei_all.rng' schematypens='http://relaxng.org/ns/structure/1.0'?><TEI xmlns="http://www.tei-c.org/ns/1.0">
	<teiHeader>
		<fileDesc>
			<titleStmt><title level='a'>Polynomial Bounds for Chromatic Number. IV: A Near-polynomial Bound for Excluding the Five-vertex Path</title></titleStmt>
			<publicationStmt>
				<publisher>Springer Link</publisher>
				<date>10/01/2023</date>
			</publicationStmt>
			<sourceDesc>
				<bibl> 
					<idno type="par_id">10517553</idno>
					<idno type="doi">10.1007/s00493-023-00015-w</idno>
					<title level='j'>Combinatorica</title>
<idno>0209-9683</idno>
<biblScope unit="volume">43</biblScope>
<biblScope unit="issue">5</biblScope>					

					<author>Alex Scott</author><author>Paul Seymour</author><author>Sophie Spirkl</author>
				</bibl>
			</sourceDesc>
		</fileDesc>
		<profileDesc>
			<abstract><ab><![CDATA[<title>Abstract</title> <p>A graph<italic>G</italic>is<italic>H</italic><italic>-free</italic>if it has no induced subgraph isomorphic to<italic>H</italic>. We prove that a<inline-formula><alternatives><tex-math>$$P_5$$</tex-math><math><msub><mi>P</mi><mn>5</mn></msub></math></alternatives></inline-formula>-free graph with clique number<inline-formula><alternatives><tex-math>$$\omega \ge 3$$</tex-math><math><mrow><mi>ω</mi><mo>≥</mo><mn>3</mn></mrow></math></alternatives></inline-formula>has chromatic number at most<inline-formula><alternatives><tex-math>$$\omega ^{\log _2(\omega )}$$</tex-math><math><msup><mi>ω</mi><mrow><msub><mo>log</mo><mn>2</mn></msub><mrow><mo>(</mo><mi>ω</mi><mo>)</mo></mrow></mrow></msup></math></alternatives></inline-formula>. The best previous result was an exponential upper bound<inline-formula><alternatives><tex-math>$$(5/27)3^{\omega }$$</tex-math><math><mrow><mrow><mo>(</mo><mn>5</mn><mo>/</mo><mn>27</mn><mo>)</mo></mrow><msup><mn>3</mn><mi>ω</mi></msup></mrow></math></alternatives></inline-formula>, due to Esperet, Lemoine, Maffray, and Morel. A polynomial bound would imply that the celebrated Erdős-Hajnal conjecture holds for<inline-formula><alternatives><tex-math>$$P_5$$</tex-math><math><msub><mi>P</mi><mn>5</mn></msub></math></alternatives></inline-formula>, which is the smallest open case. Thus, there is great interest in whether there is a polynomial bound for<inline-formula><alternatives><tex-math>$$P_5$$</tex-math><math><msub><mi>P</mi><mn>5</mn></msub></math></alternatives></inline-formula>-free graphs, and our result is an attempt to approach that.</p>]]></ab></abstract>
		</profileDesc>
	</teiHeader>
	<text><body xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xlink="http://www.w3.org/1999/xlink">
<div xmlns="http://www.tei-c.org/ns/1.0"><head n="1">Introduction</head><p>If G, H are graphs, we say G is H-free if no induced subgraph of G is isomorphic to H; and for a graph G, we denote the number of vertices, the chromatic number, the size of the largest clique, and the size of the largest stable set by |G|, &#967;(G), &#969;(G), &#945;(G) respectively.</p><p>The k-vertex path is denoted by P k , and P 4 -free graphs are well-understood; every P 4 -free graph G with more than one vertex is either disconnected or disconnected in the complement <ref type="bibr">[24]</ref>, which implies that &#967;(G) = &#969;(G). Here we study how &#967;(G) depends on &#969;(G) for P 5 -free graphs G.</p><p>The Gy&#225;rf&#225;s-Sumner conjecture <ref type="bibr">[10,</ref><ref type="bibr">25]</ref> says:</p><p>1.1 Conjecture: For every forest H there is a function f such that &#967;(G) &#8804; f (&#969;(G)) for every</p><p>This is open in general, but has been proved <ref type="bibr">[10]</ref> when H is a path, and for several other simple types of tree <ref type="bibr">([3, 11, 12, 13, 14, 17, 19]</ref>; see <ref type="bibr">[18]</ref> for a survey). The result is also known if all induced subdivisions of a tree are excluded <ref type="bibr">[17]</ref>.</p><p>A class of graphs is hereditary if the class is closed under taking induced subgraphs and under isomorphism, and a hereditary class is said to be &#967;-bounded if there is a function f such that &#967;(G) &#8804; f (&#969;(G)) for every graph G in the class (thus, the Gy&#225;rf&#225;s-Sumner conjecture says that, for every forest H, the class of H-free graphs is &#967;-bounded). Louis Esperet <ref type="bibr">[8]</ref> made the following conjecture:</p><p>Esperet's conjecture was recently shown to be false by Bria&#324;ski, Davies and Walczak <ref type="bibr">[2]</ref>. However, this raises the further question: which &#967;-bounded classes are polynomially &#967;-bounded? In particular, the two conjectures 1.1 and 1.2 would together imply the following, which is still open:</p><p>1.3 Conjecture: For every forest H, there exists c &gt; 0 such that &#967;(G) &#8804; &#969;(G) c for every H-free graph G. This is a beautiful conjecture. In most cases where the Gy&#225;rf&#225;s-Sumner conjecture has been proved, the current bounds are very far from polynomial, and 1.3 has been only been proved for a much smaller collection of forests (see <ref type="bibr">[15,</ref><ref type="bibr">20,</ref><ref type="bibr">22,</ref><ref type="bibr">23,</ref><ref type="bibr">21,</ref><ref type="bibr">5,</ref><ref type="bibr">16]</ref>). In <ref type="bibr">[23]</ref> we proved it for any P 5 -free tree H, but it has not been settled for any tree H that contains P 5 . In this paper we focus on the case H = P 5 .</p><p>The best previously-known bound on the chromatic number of P 5 -free graphs in terms of their clique number, due to Esperet, Lemoine, Maffray, and Morel <ref type="bibr">[9]</ref>, was exponential:</p><p>Here we make a significant improvement, showing a "near-polynomial" bound:</p><p>(The cycle of length five shows that we need to assume &#969;(G) &#8805; 3. Sumner <ref type="bibr">[25]</ref> showed that &#967;(G) &#8804; 3 when &#969;(G) = 2.) Conjecture 1.3 when H = P 5 is of great interest, because of a famous conjecture due to Erd&#337;s and Hajnal <ref type="bibr">[6,</ref><ref type="bibr">7]</ref>, that:</p><p>1.6 Conjecture: For every graph H there exists c &gt; 0 such that &#945;(G)&#969;(G) &#8805; |G| c for every H-free graph G. This is open in general, despite a great deal of effort; and in view of <ref type="bibr">[4]</ref>, the smallest graph H for which 1.6 is undecided is the graph P 5 . Every forest H satisfying 1.3 also satisfies the Erd&#337;s-Hajnal conjecture, and so showing that H = P 5 satisfies 1.3 would be a significant result. (See <ref type="bibr">[1]</ref> for some other recent progress on this question.)</p><p>We use standard notation throughout. When X &#8838; V (G), G[X] denotes the subgraph induced on X. We write &#967;(X) for &#967;(G[X]) when there is no ambiguity.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head n="2">The main proof</head><p>We denote the set of nonnegative real numbers by R + , and the set of nonnegative integers by Z + . Let f : Z + &#8594; R + be a function. We say</p><p>In this section we show that if a function f satisfies a certain inequality, then it is a binding function for all P 5 -free graphs. Then at the end we will give a function that satisfies the inequality, and deduce 1.5.</p><p>has a neighbour and a non-neighbour in A. It is complete to A if it is adjacent to every vertex of A. We begin with the following:</p><p>2.1 Let G be P 5 -free, and let f be a near-binding function for G. Let G be connected, and let X be a cutset of G. Then</p><p>Proof. We may assume (by replacing X by a subset if necessary) that X is a minimal cutset of G; and so G \ X has at least two components, and every vertex in X has a neighbour in V (B), for every component B of G \ X. Let B be one such component; we will prove that</p><p>), from which the result follows.</p><p>Choose v &#8712; X (this is possible since G is connected), and let N be the set of vertices in B adjacent to v. Let the components of B \ N be R 1 , . . . , R k , S 1 , . . . , S &#8467; , where R 1 , . . . , R k each have chromatic number more than f (&#8970;&#969;(G)/2&#8971;), and S 1 , . . . , S &#8467; each have chromatic number at most f (&#8970;&#969;(G)/2&#8971;). Let S be the union of the graphs S 1 , . . . , S &#8467; ; thus,</p><p>Let y &#8712; Y i . Thus, y has a neighbour in V (R i ); suppose that y is mixed on R i . Since R i is connected, there is an edge ab of R i such that y is adjacent to a and not to b. Now v has a neighbour in each component of G \ X, and since there are at least two such components, there is a vertex u &#8712; V (G) \ (X &#8746; V (B)) adjacent to v. But then u-v-y-a-b is an induced copy of P 5 , a contradiction. This proves (1).</p><p>( <ref type="formula">2</ref>)</p><p>From the minimality of I, for each i &#8712; I there exists y i &#8712; Y i such that for each j &#8712; I \ {i} we have that y i / &#8712; Y j ; and so the vertices y i (i &#8712; I) are all distinct. For each i &#8712; I choose r i &#8712; V (R i ). For all distinct i, j &#8712; I, if y i , y j are nonadjacent, then r i -y i -v-y j -r j is isomorphic to P 5 , a contradiction. Hence the vertices y i (i &#8712; I) are all pairwise adjacent, and adjacent to v; and so |I| &#8804; &#969;(G) -1.</p><p>All the vertices in N \ Y are adjacent to v, and so</p><p>Since there are no edges between any two of the graphs G[N \ Y ], R 1 , . . . , R k , their union (Z say) has clique number at most &#969;(G) -1 and so has chromatic number at most f (&#969;(G) -1). But V (B) is the union of Y, V (S) and V (Z); and so</p><p>This proves 2.1.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>2.2</head><p>Let &#8486; &#8805; 1, and let f : Z + &#8594; R + be non-decreasing, satisfying the following:</p><p>&#8226; f is a binding function for every P 5 -free graph H with &#969;(H) &#8804; &#8486;; and</p><p>Then f is a binding function for every P 5 -free graph G.</p><p>Proof. We prove by induction on |G| that if G is P 5 -free then f is a binding function for G. Thus, we may assume that G is P 5 -free and f is near-binding for G. If G is not connected, or &#969;(G) &#8804; &#8486;, it follows that f is binding for G, so we assume that G is connected and &#969;(G) &gt; &#8486;. Let us write w = &#969;(G) and m = &#8970;w/2&#8971;. If &#967;(G) &#8804; f (w) then f is a binding function for G, so we assume, for a contradiction, that:</p><p>(1) &#967;(G) &gt; f (w -1) + (w + 2)f (m).</p><p>We deduce that:</p><p>(2) Every cutset X of G satisfies &#967;(X) &gt; 2f (m).</p><p>If some cutset X satisfies &#967;(X) &#8804; 2f (m), then since &#967;(G \ X) &#8804; f (w -1) + wf (m) by 2.1, it follows that &#967;(G) &#8804; f (w -1) + (w + 2)f (m), contrary to (1). This proves (2).</p><p>(3) If P, Q are cliques of G, both of cardinality at least w/2, then G[P &#8746; Q] is connected.</p><p>Suppose not; then there is a minimal subset X &#8838; V (G) \ (P &#8746; Q) such that P, Q are subsets of different components (A, B say) of G \ X. From the minimality of X, every vertex x &#8712; X has a neighbour in V (A) and a neighbour in V (B). If x is mixed on A and mixed on B, then since A is connected, there is an edge a 1 a 2 of A such that x is adjacent to a 1 and not to a 2 ; and similarly there is an edge b 1 b 2 of B with x adjacent to b 1 and not to b 2 . But then a 2 -a 1 -x-b 1 -b 2 is an induced copy of P 5 , a contradiction; so every x &#8712; X is complete to at least one of A, B. The set of vertices in X complete to A is also complete to P , and hence has clique number at most m, and hence has chromatic number at most f (m); and the same for B. Thus, &#967;(X) &#8804; 2f (m), contrary to (2). This proves <ref type="bibr">(3)</ref>.</p><p>and the claim holds, so we may assume that &#8709; is a joint of B; let Y be a joint of B chosen with Y maximal, and let</p><p>Let N C (v) be the set of neighbours of v in V (C), and M = V (C) \ N C (v); and suppose that &#967;(M ) &gt; f (m). Let C &#8242; be a component of G[M ] with &#967;(C &#8242; ) &gt; f (m), and let Z be the set of vertices in N C (v) that have a neighbour in V (C &#8242; ). Thus, Z &#824; = &#8709;, since N C (v), V (C &#8242; ) &#824; = &#8709; and C is connected. If some z &#8712; Z is mixed on C &#8242; , let p 1 p 2 be an edge of C &#8242; such that z is adjacent to p 1 and not to p 2 ; then a-v-z-p 1 -p 2 is an induced copy of P 5 , a contradiction. So every vertex in Z is complete to V (C &#8242; ); but also every vertex in Y is complete to V (C) and hence to V (C &#8242; ), and so Y &#8746; Z is a joint of B, contrary to the maximality of Y . This proves (4). </p><p>, and f is near-binding for G) and every vertex in Y is complete to V (C), it follows that &#969;(G[Y ]) &#8804; w -m -1 &#8804; m, and so has chromatic number at most f (m) as claimed; and so &#967;(X) &gt; f (m). Consequently there is a clique P &#8838; X with cardinality w -m. The subgraph induced on the set of vertices of C complete to P has clique number at most m, and so has chromatic number at most f (m); and for each v &#8712; P , the set of vertices of C nonadjacent to v has chromatic number at most f (m) by ( <ref type="formula">4</ref>). Thus, &#967;(C) &#8804; (|P | + 1)f (m) = (w -m + 1)f (m). This proves (5).  <ref type="formula">5</ref>), and so &#967;(B) &#8804; (w -m + 2)f (m). This proves <ref type="bibr">(6)</ref>.</p><p>By <ref type="bibr">(6)</ref>, G \ N (a) has chromatic number at most (w -m + 2)f (m). But G[N (a)] has clique number at most w -1 and so chromatic number at most f (w -1); and so &#967;(G) &#8804; f (w -1) + (w -m + 2)f (m), contrary to (1). This proves 2.2. Now we deduce 1.5, which we restate:</p><p>, by a result of Sumner <ref type="bibr">[25]</ref>; if &#969;(G) = 3 then &#967;(G) &#8804; 5 &#8804; f (3), by an application of the result 1.4 of Esperet, Lemoine, Maffray, and Morel <ref type="bibr">[9]</ref>; and if &#969;(G) = 4 then &#967;(G) &#8804; 15 &#8804; f (4), by another application of 1.4. Consequently every P 5 -free graph G with clique number at most four has chromatic number at most f (&#969;(G)).</p><p>We claim that f (x -1) + (x + 2)f (&#8970;x/2&#8971;) &#8804; f (x)</p><p>for each integer x &gt; 4. If that is true, then by 2.2 with &#8486; = 4, we deduce that &#967;(G) &#8804; f (&#969;(G)) for every P 5 -free graph G, and so 1.5 holds. Thus, it remains to show that f (x -1) + (x + 2)f (&#8970;x/2&#8971;) &#8804; f (x)</p><p>for each integer x &gt; 4. This can be verified by direct calculation when x = 5, so we may assume that x &#8805; 6.</p><p>The derivative of f (x)/x 4 is (2 log 2 (x) -4)x log 2 (x)-5 , and so is nonnegative for x &#8805; 4. Consequently</p><p>for x &#8805; 5. Since x 2 (x 2 -2x -4) &#8805; (x -1) 4 when x &#8805; 5, it follows that</p><p>that is, f (x -1) + 2x + 4 x 2 f (x) &#8804; f (x), when x &#8805; 5. But when x &#8805; 6 (so that f (x/2) is defined and the first equality below holds), we have f (&#8970;x/2&#8971;) &#8804; f (x/2) = (x/2) log 2 (x/2) = (x/2) log 2 (x)-1 = (2/x)(x/2) log 2 (x) = (2/x 2 )f (x), and so f (x -1) + (x + 2)f (&#8970;x/2&#8971;) &#8804; f (x) when x &#8805; 6. This proves 2.3.</p></div></body>
		</text>
</TEI>
