DOCSLIB.ORG

Pronic Heron Mean Labeling on Special Cases of Generalized Peterson Graph P(N,K) and Disconnected Graphs

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Pronic Heron Mean Labeling on Special Cases of Generalized Peterson Graph P(N,K) and Disconnected Graphs Malaya Journal of Matematik, Vol. 9, No. 1, 262-266, 2021 https://doi.org/10.26637/MJM0901/0044 Pronic Heron Mean labeling on special cases of generalized Peterson graph P(n;k) and disconnected graphs S. Akila Devi1*, R. Sophia Porchelvi2 and T. Arjun3 Abstract A graph labeling is an assignment of integers to the vertices or edges or both subject to certain conditions. In this paper for a graph G(V;E) each of the vertices v 2 V(G) are assigned by pronic numbers. In this paper, we investigate the pronic heron mean labeling on special cases of generalized Petersen graph P(n;k). Also we described an algorithm to label the vertices for the pronic heron mean labeling for certain disconnected graphs. Keywords Pronic Heron Mean labeling, Generalized Peterson Graph P(n;k), Disconnected Graphs. AMS Subject Classification 05C10. 1,2Department of Mathematics, A.D.M College For Women, (Affiliated to Bharathidasan University), Nagapattinam, Tamil Nadu, India. 3Ideas2IT Technologies, Chennai, Tamil Nadu, India. *Corresponding author: [email protected]; [email protected]; [email protected] Article History: Received 08 January 2021; Accepted 09 February 2021 ©2021 MJM. Contents number of the form n(n + 1). Let G(p;q) be a graph with p ≥ 2.A pronic heron mean labeling[4] of a graph G is a bijec- 1 Introduction.......................................262 tion f : V(G) ! f0;2;6;12;:::; p(p + 1)g such thatp the result- f (u)+ f (v)+ f (u) f (v) 2 Main Theorems....................................263 ing edge labels obtained by f ∗(uv) = d e p 3 2.1 Special cases of P(n;k) ................ 263 ∗ f (u)+ f (v)+ f (u) f (v) or f (uv) = b 3 c for every uv 2 E(G) are 2.2 PHML on Disconnected Graphs.......... 264 all distinct. The present work is aimed to provide pronic 2.3 Algorithm for union of Path graphs Pm [ Pn ... 264 heron mean labeling of some special cases of Generalized 2.4 Algorithm for union of Path Graph and Cycle Graph Peterson graphs namely Durer Graph P(6;2), Mobius Kan- Cm [ Pn ............................ 265 tor GraphP(8;3), Dodecahedren Graph P(10;2), Desargues Graph P(10;3) and Nauru Graph P(12;5). Also we investi- 2.5 Algorithm for union of mK3, m ≥ 2 ......... 265 gate the existence of the same labeling on certain disconnected 2.6 Algorithm for union of C , n ≥ 5 and mK , m ≥ 2 265 n 3 graphs. 2.7 Algorithm for corona product of Comb Graph and Path Graph of (Pn K1) [ Pm, m;n ≥ 3 ....... 266 Definition 1.1. Generalized Peterson Graph For natural numbers n and k, where n > 2k, a Gen- 3 Conclusion........................................266 eralized Peterson graph P(n;k) is the graph whose vertex set References........................................266 is fu1;u2;:::;ung [ f v1;v2;:::;vng and its edge set is fuiui+1;uivi;vivi+k;1 ≤ i ≤ ng, where the subscript arithmetic 1. Introduction is done modulo n using the residues 0;1;2;:::;n − 1. All graphs we consider here are finite, undirected, Note: simple and connected. For standard notations and terminology ) For n = 6 and k = 2, the graph P(6;2) is said to be Durer in graph theory, we follow [1]. We refer [2] for a comprehen- Graph. sive survey on graph labeling and [3] for heron mean labeling. ) For n = 8 and k = 3, the graph P(8;3) is said to be Mobius A pronic number is a number which is the product of two Kantor Graph. Pronic Heron Mean labeling on special cases of generalized Peterson graph P(n;k) and disconnected graphs — 263/266 8 ) For n = 10 and k = 2, the graph P(10;2) is said to be (n + 5)2 + 2 i = 0 <> ∗ 2 Dodecaheran or Dodecahedral Graph. f (vivi+5) = (n + 6) + 2 i = 1: ) n = k = P( ; ) For 10 and 3, the graph 10 3 is said to be :>(n + 3)2 i = 2 Desargues Graph. ) For n = 12 and k = 5, the graph P(12;5) is said to be In the view of the above defined labeling, the Mobius Kantor Nauru Graph. graph admits pronic heron mean labeling. Theorem 2.3. Dodecahedral Graph P(10;2) is a pronic heron 2. Main Theorems mean graph. 2.1 Special cases of P(n;k) Proof. Let fv0;v1;v2;:::;v9g be the inner vertices and Theorem 2.1. Durer Graph P(6;2) admits pronic heron mean fu0;u1;u2;:::;u9g be the outer vertices of P(10;2). labeling. Define a bijection f : V(G) ! fp1; p2;:::; p2ng by 8 Proof. Let fv0;v1;v2;v3;v4;v5g be the inner vertices and p i = 2;3;:::9 ( <> i+1 fu0;u1;u2;u3;u4;u5g be the outer vertices of P(6;2). pi+10 i = 1;2:::9: f (ui) = pi+2 i = 0; ; f (vi) = Define a bijection f : V(G) ! fp1; p2;:::; p2ng by > p20 i = 0 :pi i = 1 f (u ) = p ;i = 0;1;:::;5; f (v ) = p ;i = 0;1;:::;5: i i+1 i 7+i For the above vertex labeling, the edge labeling f ∗ : E(G) ! For the above vertex labeling, the edge labeling f ∗ : E(G) ! f4;5;6;:::; p2ng is defined by f4;5;6;:::; p g is defined by ( 2n (i + 2)2 i = 0;2;3;:::;8; f ∗(u v ) = ∗ 2 i i+1 f (uivi+1) = (i + 2) ;i = 0;1;2;:::;4; (i + 1)(i + 2) i = 1 2 p 2 ∗ n + n + 2 + 2(n) + 2n 2 p 2 f (u0un−1) = d e; ∗ n + n + 6 + 6(n) + 6n 3 f (u0un−1) = d e; ∗ 3 f (vivi+2) = p8+i;i = 0;1;2;3; ( f ∗(v v ) = + i;i = ; ∗ p11+i;i = 1;2;:::;7; i i+4 91 20 0 1; f (vivi+1) = ∗ 2 p16 + 5;i = 0 f (uivi) = (i + 5) − (i + 2);i = 0;1;2;:::5: ∗ f (vivi+8) = 380 − 135i;i = 0;1; In the view of the above defined labeling, the Durer graph 8 2 admits pronic heron mean labeling. >(n + i − 4) + 6;i = 2;3;:::9 ∗ < f (uivi) = 5(i + 9);i = 1: Theorem 2.2. Mobius Kantor Graph P(8;3) admits pronic :>15n + 9;i = 0 heron mean labeling. In the view of the above defined labeling, the Dodecahedral Proof. Let fv0;v1;v2;:::;v7g be the inner vertices and graph admits pronic heron mean labeling. fu ;u ;u ;:::;u g be the outer vertices of P(8;3). 0 1 2 7 Desargues Graph P( ; ) admits pronic heron Define a bijection f : V(G) ! fp ; p ;:::; p g by Theorem 2.4. 10 3 1 2 2n mean labeling. ( pi+10 i = 0;1;2;:::;6: Proof. Let fv0;v1;v2;:::;v9g be the inner vertices and f (ui) = pi+1;i = 0;1;2;:::;7; f (vi) = pi+2 i = 7 fu0;u1;u2;:::;u9g be the outer vertices of P(10;3). Define a bijection f : V(G) ! fp0; p1; p2;:::; p2ng by ∗ For the above vertex labeling, the edge labeling f : E(G) ! 8 p i = 2;3;:::9 f4;5;6;:::; p2ng is defined by <> i+1 f (ui) = pi+2 i = 0; ∗ 2 f (uiui+ ) = (i + 2) ;i = 0;1;2:::;6; > 1 :pi i = 1 2 p 2 ( ∗ n + n + 2 + 2(n) + 2n p i = ; ::: : f (u0un−1) = d e: i+10 1 2 9 3 f (vi) = p20 i = 0 ( 2 ∗ (n + i − 2) + 6 i = 0;1;2:::;6; For the above vertex labeling, the edge labeling f ∗ : E(G) ! f (uivi) = 2 (i + 2) i = 7 f4;5;6;:::; p2ng is defined by ( 2 ( 2 ∗ (n + i + 4) i = 0;1;2;3; ∗ (i + 2) i = 0;2;3;:::;8; f (vivi+3) = f (uivi+1) = (i + 8)2 + 2 i = 4 (i + 1)(i + 2) i = 1 263 Pronic Heron Mean labeling on special cases of generalized Peterson graph P(n;k) and disconnected graphs — 264/266 2 p 2 ∗ n + n + 6 + 6(n) + 6n 2.3 Algorithm for union of Path graphs Pm [ Pn f (u0un−1) = d e; 3 The union graph Pm [ Pn where m;n ≥ 2has the ( vertex set V(Pm [ Pn)= V(Pm) [ V(Pn) with the cardinality (n + + i)2 i = ;:::; ∗ 2 1 6; m + n and edge set E(Pm [ Pn)= E(Pm) [ E(Pn) with cardinal- f (vivi+3) = (n + 7)2 + 4 i = 0 ity q = m + n − 2. Let the path Pm is demonstrated by listing the vertices and the ( 2 ∗ (n + 9) i = 0 edges in order f (vivi+7) = (n + 4 + i)2 + 4 i = 1;2; u1;e1;u2;e2;:::um−1;em−1;um. We name the vertex u1, the active vertex and the vertex um is the end vertex of the edge 8 (n + i − 4)2 + 6;i = 2;3;:::9 em−1: Now the path Pn is demonstrated by listing the ver- <> 0 0 0 ∗ tices and the edges in order v1;e ;v2;e ;:::vm−1;e ;vm. We f (uivi) = 5(i + 9);i = 1: 1 2 m−1 name the vertex v1, the first vertex and the vertex vm is the :>15n + 9;i = 0 0 end vertex of the edge em−1: In the view of the above defined labeling, the Desargues graph Label the vertices in cloclwise direction C. The algorithm admits pronic heron mean labeling. has single pass: it labels the vertices of both the paths Pm and Pn.
Load more

Recommended publications
  • Drawing Graphs and Maps with Curves
    Report from Dagstuhl Seminar 13151 Drawing Graphs and Maps with Curves Edited by Stephen Kobourov1, Martin Nöllenburg2, and Monique Teillaud3 1 University of Arizona – Tucson, US, [email protected] 2 KIT – Karlsruhe Institute of Technology, DE, [email protected] 3 INRIA Sophia Antipolis – Méditerranée, FR, [email protected] Abstract This report documents the program and the outcomes of Dagstuhl Seminar 13151 “Drawing Graphs and Maps with Curves”. The seminar brought together 34 researchers from different areas such as graph drawing, information visualization, computational geometry, and cartography. During the seminar we started with seven overview talks on the use of curves in the different communities represented in the seminar. Abstracts of these talks are collected in this report. Six working groups formed around open research problems related to the seminar topic and we report about their findings. Finally, the seminar was accompanied by the art exhibition Bending Reality: Where Arc and Science Meet with 40 exhibits contributed by the seminar participants. Seminar 07.–12. April, 2013 – www.dagstuhl.de/13151 1998 ACM Subject Classification I.3.5 Computational Geometry and Object Modeling, G.2.2 Graph Theory, F.2.2 Nonnumerical Algorithms and Problems Keywords and phrases graph drawing, information visualization, computational cartography, computational geometry Digital Object Identifier 10.4230/DagRep.3.4.34 Edited in cooperation with Benjamin Niedermann 1 Executive Summary Stephen Kobourov Martin Nöllenburg Monique Teillaud License Creative Commons BY 3.0 Unported license © Stephen Kobourov, Martin Nöllenburg, and Monique Teillaud Graphs and networks, maps and schematic map representations are frequently used in many fields of science, humanities and the arts.
    [Show full text]
  • Vertex-Transitive Graphs That Have No Hamilton Decomposition
    Vertex-transitive graphs that have no Hamilton decomposition Darryn Bryant ∗ Matthew Dean † Abstract It is shown that there are infinitely many connected vertex-transitive graphs that have no Hamilton decomposition, including infinitely many Cayley graphs of valency 6, and including Cayley graphs of arbitrarily large valency. 1 Introduction A famous question of Lov´asz concerns the existence of Hamilton paths in vertex-transitive graphs [28], and no example of a connected vertex-transitive graph with no Hamilton path is known. The related question concerning the existence of Hamilton cycles in vertex-transitive graphs is another interesting and well-studied problem in graph theory, see the survey [23]. A Hamiltonian graph is a graph containing a Hamilton cycle. Thomassen (see [10, 23]) has conjectured that there are only finitely many non-Hamiltonian connected vertex-transitive graphs. On the other hand, Babai [8, 9] has conjectured that there are infinitely many such graphs. To date only five are known. These arXiv:1408.5211v3 [math.CO] 12 Nov 2014 are the complete graph of order 2, the Petersen graph, the Coxeter graph, and the two graphs obtained from the Petersen and Coxeter graphs by replacing each vertex with a triangle. For a regular graph of valency at least 4, a stronger property than the existence of a Hamilton cycle is the existence of a Hamilton decomposition. If X is a k-valent graph, then a Hamilton k decomposition of X is a set of ⌊ 2 ⌋ pairwise edge-disjoint Hamilton cycles in X. Given the small number of non-Hamiltonian connected vertex-transitive graphs, and the uncertainty concerning the existence of others, it is natural to ask how many connected vertex-transitive graphs have no Hamilton decomposition.
    [Show full text]
  • Lombardi Drawings of Graphs 1 Introduction
    Lombardi Drawings of Graphs Christian A. Duncan1, David Eppstein2, Michael T. Goodrich2, Stephen G. Kobourov3, and Martin Nollenburg¨ 2 1Department of Computer Science, Louisiana Tech. Univ., Ruston, Louisiana, USA 2Department of Computer Science, University of California, Irvine, California, USA 3Department of Computer Science, University of Arizona, Tucson, Arizona, USA Abstract. We introduce the notion of Lombardi graph drawings, named after the American abstract artist Mark Lombardi. In these drawings, edges are represented as circular arcs rather than as line segments or polylines, and the vertices have perfect angular resolution: the edges are equally spaced around each vertex. We describe algorithms for finding Lombardi drawings of regular graphs, graphs of bounded degeneracy, and certain families of planar graphs. 1 Introduction The American artist Mark Lombardi [24] was famous for his drawings of social net- works representing conspiracy theories. Lombardi used curved arcs to represent edges, leading to a strong aesthetic quality and high readability. Inspired by this work, we intro- duce the notion of a Lombardi drawing of a graph, in which edges are drawn as circular arcs with perfect angular resolution: consecutive edges are evenly spaced around each vertex. While not all vertices have perfect angular resolution in Lombardi’s work, the even spacing of edges around vertices is clearly one of his aesthetic criteria; see Fig. 1. Traditional graph drawing methods rarely guarantee perfect angular resolution, but poor edge distribution can nevertheless lead to unreadable drawings. Additionally, while some tools provide options to draw edges as curves, most rely on straight-line edges, and it is known that maintaining good angular resolution can result in exponential draw- ing area for straight-line drawings of planar graphs [17,25].
    [Show full text]
  • Use the K-Neighborhood Subgraphs to Compute Canonical Labelings of Graphs
    mathematics Article Use the K-Neighborhood Subgraphs to Compute Canonical Labelings of Graphs Jianqiang Hao 1,*, Yunzhan Gong 2, Jianzhi Sun 1 and Li Tan 1 1 Beijing Key Laboratory of Big Data Technology for Food Safety, Beijing Technology and Business University, No. 11, Fu Cheng Road, Beijing 100048, China 2 State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, No 10, Xitucheng Road, Haidian District, Beijing 100876, China * Correspondence: [email protected]; Tel.: +86-10-6898-5704 Received: 5 July 2019; Accepted: 27 July 2019; Published: 31 July 2019 Abstract: This paper puts forward an innovative theory and method to calculate the canonical labelings of graphs that are distinct to Nauty’s. It shows the correlation between the canonical labeling of a graph and the canonical labeling of its complement graph. It regularly examines the link between computing the canonical labeling of a graph and the canonical labeling of its open k-neighborhood subgraph . It defines dif fusion degree sequences and entire dif fusion degree sequence . For each node of a graph G, it designs a characteristic m_NearestNode to improve the precision for calculating canonical labeling. Two theorems established here display how to compute the first nodes of MaxQ(G). Another theorem presents how to determine the second nodes of MaxQ(G). When computing Cmax(G), if MaxQ(G) already holds the first i nodes u1, u2, ··· , ui, Diffusion and Nearest Node theorems provide skill on how to pick the succeeding node of MaxQ(G). Further, it also establishes two theorems to determine the Cmax(G) of disconnected graphs.
    [Show full text]
  • Math.RA] 25 Sep 2013 Previous Paper [3], Also Relying in Conceptually Separated Tools from Them, Such As Graphs and Digraphs
    Certain particular families of graphicable algebras Juan Núñez, María Luisa Rodríguez-Arévalo and María Trinidad Villar Dpto. Geometría y Topología. Facultad de Matemáticas. Universidad de Sevilla. Apdo. 1160. 41080-Sevilla, Spain. [email protected] [email protected] [email protected] Abstract In this paper, we introduce some particular families of graphicable algebras obtained by following a relatively new line of research, ini- tiated previously by some of the authors. It consists of the use of certain objects of Discrete Mathematics, mainly graphs and digraphs, to facilitate the study of graphicable algebras, which are a subset of evolution algebras. 2010 Mathematics Subject Classification: 17D99; 05C20; 05C50. Keywords: Graphicable algebras; evolution algebras; graphs. Introduction The main goal of this paper is to advance in the research of a novel mathematical topic emerged not long ago, the evolution algebras in general, and the graphicable algebras (a subset of them) in particular, in order to obtain new results starting from those by Tian (see [4, 5]) and others already obtained by some of us in a arXiv:1309.6469v1 [math.RA] 25 Sep 2013 previous paper [3], also relying in conceptually separated tools from them, such as graphs and digraphs. Concretely, our goal is to find some particular types of graphicable algebras associated with well-known types of graphs. The motivation to deal with evolution algebras in general and graphicable al- gebras in particular is due to the fact that at present, the study of these algebras is very booming, due to the numerous connections between them and many other branches of Mathematics, such as Graph Theory, Group Theory, Markov pro- cesses, dynamic systems and the Theory of Knots, among others.
    [Show full text]
  • A Godsil E Thomason I Solomon M Abiad B Van Dam F Balbuena J
    8.30–9.15 am A Godsil E Thomason I Solomon M Abiad 9.20–9.40 am 1 Zhan 10 Bukh 19 Srinivasan 24 Reichard 9.45–10.05 am 2 Ye 11 Martin 20 Sumalroj 25 Xu 10.05–10.35 am break break break break 10.35–10.55 am 3 Dalfó 12 Kamat 21 Bencs 26 Peng 11.00–11.20 am 4 McGinnis 13 Timmons 22 Guo 27 Kravitz 11.25–12.10 pm B Van Dam F Balbuena J Muzychuk N Xiang 12.10–2.00 pm lunch lunch lunch lunch 2.00–2.45 pm C Kantor G Füredi K Williford 2.50–3.10 pm 5 Gu 14 Tait 23 Ducey 3.15–3.35 pm 6 Coutinho 15 Kodess L Woldar break 3.40–4.00 pm 7 Greaves 16 Y. Wang 4.00–4.30 pm break break 4.30–4.50 pm 8 Fiol 17 Kronenthal 4:55–5.15 pm 9 W. Wang 18 Moorhouse 5.20–6.10 pm D Haemers H Lazebnik 2 8.30–9.15 am A Chris Godsil Spectral invariants from embeddings 9.20–9.40 am 1 Harmony Zhan Quantum walks and mixing 9.45–10.05 am 2 Dong Ye Median eigenvalues and graph inverse 10.35–10.55 am 3 Cristina Dalfó Characterizing identifying codes from the spectrum of a graph or digraph 11.00–11.20 am 4 Matt McGinnis The smallest eigenvalues of the Hamming graphs 11.25–12:10 pm B Edwin van Dam Partially metric association schemes with a small multiplicity 2.00–2.45 pm C Bill Kantor MUBs 2.50–3.10 pm 5 Xiaofeng Gu Toughness, connectivity and the spectrum of regular graphs 3.15–3.35 pm 6 Gabriel Coutinho Average mixing matrix 3.40–4.00 pm 7 Gary Greaves Edge-regular graphs and regular cliques 4.30–4.50 pm 8 Miguel Angel Fiol An algebraic approach to lifts of digraphs 4.55–5.15 pm 9 Wei Wang A positive proportion of multigraphs are determined by their generalized spectra 5.20–6.10
    [Show full text]
  • The Cost of Perfection for Matchings in Graphs
    The Cost of Perfection for Matchings in Graphs Emilio Vital Brazil∗ Guilherme D. da Fonseca† Celina M. H. de Figueiredo‡ Diana Sasaki‡ September 15, 2021 Abstract Perfect matchings and maximum weight matchings are two fundamental combinatorial structures. We consider the ratio between the maximum weight of a perfect matching and the maximum weight of a general matching. Motivated by the computer graphics application in triangle meshes, where we seek to convert a triangulation into a quadrangulation by merging pairs of adjacent triangles, we focus mainly on bridgeless cubic graphs. First, we characterize graphs that attain the extreme ratios. Second, we present a lower bound for all bridgeless cubic graphs. Third, we present upper bounds for subclasses of bridgeless cubic graphs, most of which are shown to be tight. Additionally, we present tight bounds for the class of regular bipartite graphs. 1 Introduction The study of matchings in cubic graphs has a long history in combinatorics, dating back to Pe- tersen’s theorem [20]. Recently, the problem has found several applications in computer graphics and geographic information systems [5, 18, 24, 10]. Before presenting the contributions of this paper, we consider the following motivating example in the area of computer graphics. Triangle meshes are often used to model solid objects. Nevertheless, quadrangulations are more appropriate than triangulations for some applications [10, 23]. In such situations, we can convert a triangulation into a quadrangulation by merging pairs of adjacent triangles (Figure 1). Hence, the problem can be modeled as a matching problem by considering the dual graph of the triangulation, where each triangle corresponds to a vertex and edges exist between adjacent triangles.
    [Show full text]
  • On the Euclidean Dimension of Graphs
    ON THE EUCLIDEAN DIMENSION OF GRAPHS JIN HYUP HONG GREAT NECK SOUTH HIGH SCHOOL, GREAT NECK, NY AND DAN ISMAILESCU MATHEMATICS DEPARTMENT, HOFSTRA UNIVERSITY, NY arXiv:1501.00204v1 [math.MG] 31 Dec 2014 1 Abstract. The Euclidean dimension a graph G is defined to be the smallest integer d such that the vertices of G can be located in Rd in such a way that two vertices are unit distance apart if and only if they are adjacent in G. In this paper we determine the Euclidean dimension for twelve well known graphs. Five of these graphs, D¨urer, Franklin, Desargues, Heawood and Tietze can be embedded in the plane, while the remaining graphs, Chv´atal, Goldner-Harrary, Herschel, Fritsch, Gr¨otzsch, Hoffman and Soifer have Euclidean dimension 3. We also present explicit embeddings for all these graphs. 1. History and previous work The Euclidean dimension of a graph G = (V, E), denoted dim(G) is the least integer n such that there exists a 1 : 1 embedding f : V Rn for which f(u) f(v) = 1 if and only → | − | if uv E. ∈ The concept was introduced by Erd˝os, Harary and Tutte in their seminal paper [7], where the authors determine the Euclidean dimension for several classes of graphs. For instance, they show that dim(K ) = n 1, where K is the complete graph on n n − n vertices. Using a construction due to Lenz, they also compute the Euclidean dimension of Km,n, the complete bipartite graph with m vertices in one class and n vertices in the other.
    [Show full text]
  • Graph Neural Network
    Graph Networks for Track Reconstruction Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 Marcel Kunze, Heidelberg University Data Representation Numeric: made of numbers ● Plain vector ● Sequences Categorical: made of words ● Trees Female ● Graph 1 0 Male 0 1 The numbers do not represent a complex relationship or hierarchy Graph Networks - Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 | M.Kunze 2 Data Representation ● Plain vectors User events sequence ● Sequences ● Trees ● Graph The data objects have a sequential (time) dependence Graph Networks - Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 | M.Kunze 3 Data Representation ● Plain vectors ● Sequences ● Trees ● Graph The data objects have a hierarchical dependence Graph Networks - Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 | M.Kunze 4 Data Representation ● Plain vectors ● Sequences ● Trees ● Graph The data objects have an arbitrary dependence • Model complex relationship / hierarchy Graph Networks - Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 | M.Kunze 5 Directed Graphs A directed Graph is a graph oF nodes whose edges are all E directed Applications D • one-way streets • flights C • task scheduling • … B A Graph Networks - Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 | M.Kunze 6 Directed Acyclic Graphs (DAG) A directed acyclic graph or DAG is a directed graph with no directed cycles: Graph Networks - Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 | M.Kunze 7 Gaming: Sparse Voxel Octrees (SVO) 1 2 3 4 5 • Raytracing • Compression of data • Multi-scale resolution Graph Networks - Learning to Discover: Advanced Pattern Recognition, IPa Workshop 2019 | M.Kunze 8 Adjacency Matrix • The adjacency matrix is a square matrix to represent a graph.
    [Show full text]
  • Arxiv:2008.08856V2 [Math.CO] 5 Dec 2020 Graphs Correspond Exactly to the Graphs Induced by the Two Middle Layers of Odd-Dimensional Hypercubes
    Fast recognition of some parametric graph families Nina Klobas∗ and Matjaž Krnc† December 8, 2020 Abstract We identify all [1, λ, 8]-cycle regular I-graphs and all [1, λ, 8]-cycle reg- ular double generalized Petersen graphs. As a consequence we describe linear recognition algorithms for these graph families. Using structural properties of folded cubes we devise a o(N log N) recognition algorithm for them. We also study their [1, λ, 4], [1, λ, 6] and [2, λ, 6]-cycle regularity and settle the value of parameter λ. 1 Introduction Important graph classes such as bipartite graphs, (weakly) chordal graphs, (line) perfect graphs and (pseudo)forests are defined or characterized by their cycle structure. A particularly strong description of a cyclic structure is the notion of cycle-regularity, introduced by Mollard [24]: For integers l, λ, m a simple graph is [l, λ, m]-cycle regular if ev- ery path on l+1 vertices belongs to exactly λ different cycles of length m. It is perhaps natural that cycle-regularity mostly appears in the literature in the context of symmetric graph families such as hypercubes, Cayley graphs or circulants. Indeed, Mollard showed that certain extremal [3, 1, 6]-cycle regular arXiv:2008.08856v2 [math.CO] 5 Dec 2020 graphs correspond exactly to the graphs induced by the two middle layers of odd-dimensional hypercubes. Also, for [2, 1, 4]-cycle regular graphs, Mulder [25] showed that their degree is minimized in the case of Hadamard graphs, or in the case of hypercubes. In relation with other graph families, Fouquet and Hahn [12] described the symmetric aspect of certain cycle-regular classes, while in [19] authors describe all [1, λ, 8]-cycle regular members of generalized ∗Department of Computer Science, Durham University, UK.
    [Show full text]
  • Arxiv:2008.11296V2 [Math.CO] 4 Mar 2021 8 4 8 2 8 4 8
    SEQUENCES OF WELL-DISTRIBUTED VERTICES ON GRAPHS AND SPECTRAL BOUNDS ON OPTIMAL TRANSPORT LOUIS BROWN Abstract. Given a graph G = (V; E), suppose we are interested in select- n ing a sequence of vertices (xj )j=1 such that fx1; : : : ; xkg is `well-distributed' uniformly in k. We describe a greedy algorithm motivated by potential the- ory and corresponding developments in the continuous setting. The algorithm performs nicely on graphs and may be of use for sampling problems. We can interpret the algorithm as trying to greedily minimize a negative Sobolev norm; we explain why this is related to Wasserstein distance by establishing a purely spectral bound on the Wasserstein distance on graphs that mirrors R. Peyre's estimate in the continuous setting. We illustrate this with many examples and discuss several open problems. 1. Introduction 1.1. Introduction. The purpose of this paper is to discuss a basic problem on finite graphs G = (V; E) that is already difficult on the unit interval [0; 1]. Given a metric space (say, the unit interval, a compact manifold or a finite graph), how does one construct a sequence x1; x2;::: of elements in the space such that their distribution is uniformly good|by this we mean that if one takes the first k el- ements fx1; : : : ; xkg, then this set is very nearly as evenly distributed on the set as any set of k elements would be. We have not clearly defined what notion of `well-distributed' we mean, this will depend on the actual setting; the question is frequently interesting for several different such notions.
    [Show full text]
  • Matching Preclusion of the Generalized Petersen Graph
    Theory and Applications of Graphs Volume 6 Issue 1 Article 5 May 2019 Matching Preclusion of the Generalized Petersen Graph Ajay Arora Novi High School, [email protected] Eddie Cheng Oakland University, [email protected] Christopher Melekian Oakland University, [email protected] Follow this and additional works at: https://digitalcommons.georgiasouthern.edu/tag Part of the Discrete Mathematics and Combinatorics Commons Recommended Citation Arora, Ajay; Cheng, Eddie; and Melekian, Christopher (2019) "Matching Preclusion of the Generalized Petersen Graph," Theory and Applications of Graphs: Vol. 6 : Iss. 1 , Article 5. DOI: 10.20429/tag.2019.060105 Available at: https://digitalcommons.georgiasouthern.edu/tag/vol6/iss1/5 This article is brought to you for free and open access by the Journals at Digital Commons@Georgia Southern. It has been accepted for inclusion in Theory and Applications of Graphs by an authorized administrator of Digital Commons@Georgia Southern. For more information, please contact [email protected]. Matching Preclusion of the Generalized Petersen Graph Cover Page Footnote This paper originated from a project started at the Oakland University Summer Mathematics Institute. This article is available in Theory and Applications of Graphs: https://digitalcommons.georgiasouthern.edu/tag/vol6/ iss1/5 Arora et al.: Matching Preclusion of the Generalized Petersen Graph Abstract The matching preclusion number of a graph with an even number of vertices is the minimum number of edges whose deletion results in a graph with no perfect matchings. In this paper we determine the matching preclusion number for the generalized Petersen graph P (n; k) and classify the optimal sets. Keywords: perfect matching; Petersen graph; generalized Petersen graph.
    [Show full text]