A New Spectral Technique Using Normalized Adjacency Matrices for Graph Matching 1

Size: px
Start display at page:

Download "A New Spectral Technique Using Normalized Adjacency Matrices for Graph Matching 1"

Transcription

1 CHAPTER-3 A New Spectral Technique Using Normalized Adjacency Matrices for Graph Matching Graph matching problem has found many applications in areas as diverse as chemical structure analysis, pattern recognition and bioinformatics. Many works have been reported for matching of specialized types of graphs. A new spectral technique based on eigenvalues and minimum eigenvector of the normalized adjacency matrix of the graphs is proposed in this chapter. The adjacency matrix representation of the graph is combined with the explicit information about the degree of each vertex to generate the normalized adjacency matrix of the graph [Hogben, 25]. The spectral properties of this matrix and the shortest distance sum from a vertex to other vertices are employed to match graphs and verify graph similarity/ isomorphism. Further the correspondence of the vertices between graphs is established by matching the values in the minimum eigenvectors of the normalized adjacency matrices. 3. Introduction In literature different matrix representations of the graphs are often used for finding the spectral properties. The matrix representations, which are employed for the purpose are, the Adjacency matrices, Laplacian matrices, the Signless Laplacian matrices and their normalized forms. In this work the normalized adjacency matrix representation of the graph and its spectral characteristics are employed for matching of the graphs along with the shortest distance sum to other vertices and vertex degrees. Further the eigenvalues and the minimum eigenvectors of such matrices representing the two graphs are used for finding vertex correspondence. Parts of this chapter appear in the paper Spectral Technique using Normalized Adjacency Matrices for Graph Matching published in International Journal of Computational Science and Mathematics, ISSN , Vol. 3 No. 4 (2), pp

2 The spectra of various matrices (i.e., the eigenvalues of the matrices) are used to obtain information about a graph. Spectral graph theory deals with exploration of graphs and study of their properties in relationship to the characteristic polynomial, eigenvalues, and eigenvectors of matrices associated with the graph, such as its adjacency matrix or laplacian matrix or their normalized forms. Two graphs are called isospectral or cospectral if the representing matrices of the graphs have equal multisets of eigenvalues [W]. The characteristic graph spectral properties of the various matrix representations are described in the following. The most common matrices that have been studied for simple undirected and unweighted graphs are; The Adjacency Matrix The Laplacian (Combinatorial) Matrix The Normalized Laplacian The Signless Laplacian The Normalized Adjacency Matrix 3.. The Adjacency Matrix Let G be a simple graph, one can define a matrix X(G), called the adjacency matrix, This representation is for undirected graph which implies that the adjacency matrix is symmetric. The spectrum of such a graph includes a set of eigenvalues which may be repeated and real. Such a matrix has a full set of eigenvectors which are mutually orthogonal. 53

3 The adjacency matrix uses s and s. This is useful for emphasizing the discrete nature of the graph. Other values can be assigned if it is a weighted graph. Some other spectral properties of the adjacency matrix representation of undirected graphs are listed below; The sum of the eigenvalues is the trace of the adjacency matrix which is since X(A) has for all elements on the diagonal. The sum of the square of the eigenvalues is the same as the trace of X(A 2 ). The diagonal entries of X(A 2 ) counts the number of closed walks of length 2, for which each edge is counted exactly twice The sum of the eigenvalues cubed is the same as the trace of X(A 3 ) i.e., the same as the number of closed walks of length 3. In adjacency matrix, eigenvector if and only if the graph is regular. is an 3..2 The Laplacian Matrix (Combinatorial Matrix) The Laplacian matrix L, is also called the admittance matrix or Kirchhoff matrix or combinatorial, of a graph G, and is defined for an undirected, unweighted graph without loops or multiple edges (i.e. Simple Graph). Let V be the vertex set and E be the edge set of graph G, if X(G) is a equation 3.[W2] symmetric adjacency matrix then L is defined by (3.) Where is the diagonal degree matrix, which is the diagonal matrix formed from the vertex degrees and X(G) is the adjacency matrix. The elements of L can be obtained as given in equation 3.2; (3.2) 54

4 A few spectral properties of the Laplacian Matrix are listed below, One special eigenvalue is zero. The fact is that the sum of all the rows is zero for zero eigenvalues. All the other eigenvalues are nonnegative, so the combinatorial Laplacian is positive semi-definite. For a graph G and its Laplacian matrix L with eigenvalues L is always positive semi-definite. Relation between Laplacian matrix L and incidence matrix M : If we define an oriented incidence matrix M with element M ev for edge e (connecting vertex i and j, with i < j) and vertex v is given by equation 3.3. (3.3) Then the laplacian matrix L satisfies the realtion, Where is the matrix transpose of The second smallest eigenvalue of the Laplacian matrix of G is the algebraic connectivity of G. This eigenvalue is named the algebraic connectivity of the graph because it serves as a lower bound on the degree of robustness of the graph to node and edge failures. This follows from the following inequality where, ν(g) is the node-connectivity and η (G) is the edge-connectivity of a graph. Therefore, a network with high algebraic connectivity is robust to both node and edge failures. The algebraic connectivity of a graph G is greater than if and only if G is a connected graph. The value of the algebraic connectivity is bounded above by the vertex connectivity of the graph. The algebraic connectivity is bounded below 55

5 by, where n is the number of vertices of a connected graph and D is the diameter and in fact in a result due to Brendan McKay by [W6]. Unlike the traditional connectivity, the algebraic connectivity is dependent on the number of vertices, as well as the way in which vertices are connected. In random graphs, the algebraic connectivity decreases with the number of vertices, and increases with the average degree. The number of times zero appears as an eigenvalue in the Laplacian gives the number of connected components in the graph. The smallest non-zero eigenvalue of L is called the spectral gap. These are some important spectral properties of the Laplacian matrix The Normalized Laplacian Matrix A normalized version of the Laplacian matrix, denoted by equation 3.4 [ Hall, 2] is formally defined as in if i jand d L i, j - if i and j are adjacent (3.4) d d i j otherwise j For a graph G with no isolated vertices the relationship between normalized Laplacian and the Laplacian matrix L is given by equation 3.5 Some of the spectral properties of the normalized laplacian are enlisted below, Zero is an eigenvalue and the remaining eigenvalues are nonnegative. (3.5) 56

6 The difference between the spectrums of the laplacian and the normalized laplacian is that; for the combinatorial laplacian the eigenvalues can be essentially large, whereas the normalized laplacian has eigenvalues always lying in the range between zero and two The Signless Laplacian Matrix The Signless Laplacian matrix of a simple graph G is defined as in equation 3.6 (3.6) X(G) is a symmetric adjacency matrix and D is the diagonal matrix of vertex degrees 3..5 Normalized Adjacency Matrix The normalized adjacency matrix N(G) of graph is a unique representation that combines the degree information of each vertex and their adjacency information in the graph. The normalized adjacency matrix is obtained from the adjacency matrix of the graph. Consider an undirected graph G = (V, E) where V is the set of vertices, E is the set of edges and X(G) is the adjacency matrix representing G. The diagonal degree matrix, represents the degrees of each vertex of the graph. The normalized adjacency matrix is given by equation 3.7, N ( G) D X ( G) D, where diag( degg,... deg n). D G (3.7) N (G) is referred to as normalized adjacency matrix [W3]. The normalized adjacency matrix is defined only for graphs without isolated vertices. This is because the degree of the isolated vertex is zero and cannot be uniquely determined for such graphs. Hence the proposed methodology is suitable only for graphs without isolated vertices. 57

7 Some common spectral properties of various matrices are listed below; For an undirected graph G, all the eigenvalues of the adjacency matrix are zero if and only if G has no edges. The same holds for the Laplace eigenvalues and the signless Laplace eigenvalues. If d is a diameter of a connected graph G. Then G has at least distinct eigenvalues for an adjacency matrix. The same holds for the Laplace eigenvalues and the signless Laplace eigenvalues. Let G be a graph with connected components. Then the spectrum of G is the union of the spectra of (and multiplicities are added) for the adjacency matrix. The same holds for the Laplace and the signless Laplace spectrum. For an undirected, regular graph G of degree k. The largest eigenvalue of G is k, and its multiplicity equals the number of connected components of G in an adjacency matrix. The spectrum of J is n (with multiplicity ), zero (with multiplicity n-). Where J is all- matrix of order n and rank, and that the all- vector is an eigenvector with eigenvalue n. The spectrum of adjacency matrix A is (n ) (with multiplicity ), ( ) (with multiplicity (n-), where is an adjacency matrix of the complete graph on n vertices of graph G. The Laplace matrix is, which has spectrum zero (with multiplicity ), n (with multiplicity n-). For an undirected graph G the multiplicity of zero as Laplace eigenvalues, equals the number of connected components of G. A graph G is bipartite if and only if the Laplace spectrum and the signless Laplace spectrum of G are equal. 58

8 In this work the spectral property correspondence of graphs, along with shortest distance sum and degree invariance, has been used to prove the similarity of graphs (for graph matching). The remaining part of the chapter describes the methodology employed for finding whether two graphs are same/ similar/ isomorphic Overview of the Proposed Methodology The proposed methodology involves finding the normalized adjacency matrix representation of the two graphs and computing the eigenvalues and eigenvectors of the same. Further to find isomorphism, the degree and the shortest distance from a vertex to all the other vertices in the graph are computed. If the degree and the shortest distances sum of the two graphs are equivalent and the eigenvalues of the normalized adjacency matrices of the two graphs are the same, then the two graphs are similar/ isomorphic. Further the vertex correspondence is obtained using the equivalent values of minimum eigenvector of the normalized adjacency matrix. The methodology has been implemented using MATLAB and tested on a large number of graphs. The results are excellent. The complete methodology is described in section The Proposed Methodology Graph matching involves finding out whether two graphs are same or similar or isomorphic. Two graphs are said to be isomorphic if there is a correspondence between their vertex sets such that they also preserve adjacency. Thus is isomorphic to if there is a bijection (3.8) Clearly, isomorphic graphs have same number of vertices and edges. The diagonal degree matrix and its inverse are employed to find normalized adjacency matrix from the adjacency matrix of the graph as depicted in equation 3.7. If the 59

9 eigenvalues of the normalized adjacency matrices of the two graphs have a one to one correspondence and the shortest distance sums along with vertex degrees match then the two graphs are isomorphic. Further the eigenvectors corresponding to minimum eigenvalues of the two normalized adjacency matrices are employed for finding the mapping between the vertices of the two graphs. The eigenvalues of the normalized adjacency matrices are known to be invariant for isomorphic transformations. Hence eigenvalues of the normalized adjacency matrices of two isomorphic/ similar graphs are required to be the same. Further based on theorem 2. the following observation 3. of the spectral property of normalized adjacency matrices along with the invariance of the vertex degree and shortest distance sum of vertex with other vertices can be made. Observation 3. The invariance of vertex degree, shortest distance sum of vertices and eigenvalues of normalized adjacency matrix representation of the two graphs are sufficient conditions for graphs to be isomorphic Let N (G ), be the normalized adjacency matrix of G and N 2 (G 2 ) be the normalized adjacency matrix of the graph G 2. The spectral properties of the graphs are found by computing the eigenvalues and eigenvectors of N and N 2. Let E =Eigenvalue (N ) and E 2 =Eigenvalue (N 2 ), be the eigenvalues of the normalized adjacency matrices N and N 2. If (all the elements of E are also present in E 2 ) then the two graphs G and G 2 are candidates for similarity/ isomorphism. The isomorphism of the graphs G and G 2 is ascertained if and only if; there is equivalence of shortest distance sums and degrees of corresponding vertices between the two graphs. 6

10 Further to find vertex/ node correspondences between two isomorphic graphs, the minimum eigenvectors are employed. Let EV =Eigenvector(N ) and EV 2 =Eigenvector(N 2 ) be the eigenvectors of N and N 2. The correspondence of the vertices is found by matching the values of the eigenvectors corresponding to the minimum eigenvalues of G and G 2. If the matching results in less than matches (empirical value) one of the minimum eigenvector is negated and further matching is obtained and this gives the correspondence between the vertices of the two graphs. This process is based on the property of the graph spectra given as Observation 3.2. The complete process of the graph matching is depicted in the flow chart given in Figure 3.. The same is explained using an example in section Observation 3.2: The matching of the absolute values of the coefficients of minimum eigenvector of the normalized adjacency matrix of two isomorphic graphs, gives a possible correspondence between vertices of the two graphs. 6

11 Start Read the edge list and generate adjacency matrices X (G ) and X (G 2 ) Find the diagonal matrices of the two graphs (D,D 2 ), D= diag(d,d 2, d n ) & shortest distance sums(sp,sp 2 ) Find the normalized adjacency matrix N and N 2 using the formula D X ( G) D Flag check for shortest path & degree match bet n vertices of G and G 2, if so assign else assign Find eigenvalues of N and N 2, E = Eig(N ) and E 2 =Eig(N 2 ) Yes If E ΞE 2 & flag A No Print graphs are not isomorphic stop Figure 3.: Flow chart for graph matching using normalized adjacency matrices (continued ) 62

12 A EV = Min eigenvector (N ) EV 2 = Min eigenvector (N 2 ) i= i=i+ j= i=i+ No Yes No i n No Yes J > n EV (i)=ev 2 ( j) No Yes Map (k,) Map (k,2) k k+ i j i n Yes j j+ Print The graphs are isomorphic and MAP gives vertex correspondence No k < n/2 Yes EV2 = - EV2 Stop Figure 3.: Flow chart for graph matching using normalized adjacency matrices. 63

13 3.2. An Example The methodology for matching graphs using eigenvalues and eigenvectors of the normalized adjacency matrix is described using an example in this sub section. The example considers two six vertex graphs. The two six vertex graphs G and G 2 are shown in Figure 3.2 below V V V 5 V 6 V 2 V 5 V 3 V 2 V 3 V 6 V 4 V 4 Graph G Graph G 2 Figure 3.2: Graphs G and G 2 The adjacency matrices of G and G 2 are given below. X(G ) X(G 2 ) The normalized adjacency matrices N and N 2 are computed and given below. 64

14 N =NAM(G ) = N 2 =NAM(G 2 ) = The eigenvalues of the two graphs computed from N and N 2 and are shown below; E = E 2 = The eigenvectors of the graphs are again computed from N and N 2 and are shown below; 65

15 EV = EV 2 = The Table 3. below lists the parameters degrees, shortest distance sum of the vertex and absolute eigenvalues, the Sl No also represents the vertex number for graphs G and G 2. Table: 3.: Eigenvalues, degrees and average shortest distance of the vertices Sl.No. Degree of Vertices in G Shortest distance sum with other vertices of G Degree of Vertices in G 2 Shortest distance sum with other vertices of G 2 Eigen values of graph G Eigen values of graph G From the table 3., one can see that the shortest distance sums and degrees are equivalent (ie. the shortest distance sum and degrees with same multiplicities are found in both graphs). Further the eigenvalues are also equivalent; hence the graphs G and G 2 are isomorphic. The vertex correspondence between the two graphs is established using minimum eigenvector of N and N 2.The minimum eigenvectors of N and N 2, namely MinEig and Min Eig2 are enlisted below. 66

16 MinEig= [ ] MinEig2= [ ] Mapping of the vertices are obtained from the above vectors and are as shown below. Vertex Correspondence = Hence vertex correspondence between graphs G and G 2 are [-, 2-4, 3-3, 4-2, 5-6, 6-5], which is true by observation Comments on Correctness of the Methodology In this chapter, it has been shown that the two graphs are similar/ isomorphic, if the graphs have same number of vertices, degrees, and eigenvalues along with the shortest distance sum from a vertex to all the other vertices. A few examples showing that the graphs having the same spectra (i.e. they are cospectral) but are non-isomorphic graphs are enunciated in the following. The proposed methodology uses the following three properties for establishing vertex correspondence viz: i) Cospectral normalized adjacency matrices i.e., NAM S of two graphs N and N 2 have same eigenvalues. ii) Equivalent diagonal degree matrices. The diagonal degree matrices D and D 2 of both the graphs have equal number of similar degreed vertices. 67

17 iii) Equivalent shortest distance sum of the two graphs. The vertices with same degree have the same shortest distance sum to other vertices in both the graphs. The methodology has been tested on large number of synthetic graph pairs and accurate results have been obtained. It is known that some non-isomorphic graphs are cospectral [Dam and Haemers 23]. In the methodology along with cospectral nature of the normalized adjacency matrices of the graphs the two other characteristic features namely equivalent degrees and shortest distance sums are used. These characteristics filter out cospectral non-isomorphic graphs. For example the graphs in Figure 3.3 are cospectral and are of the same number of vertices and equivalent degrees as both of them are r-regular (r=4) graphs. They do not satisfy the equivalent shortest distance characteristic and hence are not isomorphic according to the methodology, which is true. Graph G Graph G 2 Figure 3.3: Two cospectral regular graphs G & G 2 which are non isomorphic The graph pair in figure 3.4 are also cospectral but non-isomorphic. These graphs are the smallest pair of graphs which are co-spectral. The proposed methodology correctly identifies them as non-isomorphic as they do not have equivalent degrees. 68

18 Figure 3.4: Cospectral smallest graphs The same characteristic requirement also identifies the pair of graphs in figure 3.5 as nonisomorphic. The two graphs in figure 3.5 are cospectral but are non-isomorphic as is apparent by the appearance of the two graphs. One of the graphs in figure 3.5 has a vertex with degree 5 whereas the other does not. Figure 3.5: Two non-isomorphic cospectral graphs that does not satisfy degree invariance Many such counter example graphs have been tested for matching and the results obtained are as expected. Typical graphs for isomorphism testing enlisted in chapter 2, have also obtained accurate results. Basically the shortest distance sum catches all the typically non-isomorphic graphs which look or seem isomorphic and hence is a powerful and sufficient characteristic for verifying graph matching as discussed in chapter 2. Hence the methodology proposed in this chapter is correct for undirected simple graphs. The experimentation conducted with this methodology is described in section

19 3.3 Experimentation The proposed methodology for the graph matching is implemented using MATLAB 7. The methodology has been tested on one hundred and twenty five synthetic graph pairs and % accurate results have been achieved. Among these seventy four pairs were isomorphic and the remaining graph pairs were non isomorphic. The test results for a few of the different graphs having different number of vertices have been enlisted in Table 3.2, along with the time taken by the methodology. Table 3.2: The results of graph matching using normalized adjacency matrices Sl No. of No. of Isomorphic Correct Mapping of Time for No Vertices Edges (Y/ N) Vertices (Y/ N) Computation 4 4 Y Y Y Y Y Y Y Y N Y Y Y Y Y Y N Y Y Y Y N Y Y Y Y N Y Y N Y Y Y Y Y Y Y Y N Y Y Y Y N Y Y N - - 7

20 Time in Seconds 28 3 Y Y Y Y Y Y Y Y N Y Y N Y Y Y Y Y Y Y Y Y Y Y Y The time taken by the methodology v/s the number of vertices of the graphs is plotted in Figure Computation Time Computation Time Number of nodes Figure 3.6: Time v/s Number of Vertices Plot The nature of the graph shows an increase in time with increase in the number of vertices. The time complexity of the algorithm is found to be polynomial and detailed analysis is provided in Chapter 6. 7

21 3.4 Summary In the present work a new spectral technique using normalized adjacency matrices for graph matching is presented. The method makes use of eigenvalues of the two normalized adjacency matrices of the graphs. If the vertices of the two graphs have the same eigenvalues then the two graphs are assumed to be candidates for similarity/isomorphism. The similarity is ascertained by checking for the equivalence of degrees and shortest distance sum. Further, if the graphs are found to be isomorphic/ similar, the eigenvectors corresponding to minimum eigenvalues are employed for finding the correspondence between the vertices of the graph. The program developed in MATLAB 7, is tested using about hundred and twenty five pairs of synthetic graphs with different numbers of vertices and accurate results have been achieved. This methodology has proved to be correct and efficient for a large number of graphs. Further the use of normalized adjacency matrix representation of the graph has proven to be beneficial as it contains not only the adjacency information but the degree information of each vertex stated explicitly. This representation is also simple for computation purpose and holds a lot of promise for other applications. 72

This section is an introduction to the basic themes of the course.

This section is an introduction to the basic themes of the course. Chapter 1 Matrices and Graphs 1.1 The Adjacency Matrix This section is an introduction to the basic themes of the course. Definition 1.1.1. A simple undirected graph G = (V, E) consists of a non-empty

More information

BOUNDS FOR LAPLACIAN SPECTRAL RADIUS OF THE COMPLETE BIPARTITE GRAPH

BOUNDS FOR LAPLACIAN SPECTRAL RADIUS OF THE COMPLETE BIPARTITE GRAPH Volume 115 No. 9 017, 343-351 ISSN: 1311-8080 (printed version); ISSN: 1314-3395 (on-line version) url: http://www.ijpam.eu ijpam.eu BOUNDS FOR LAPLACIAN SPECTRAL RADIUS OF THE COMPLETE BIPARTITE GRAPH

More information

1.10 Matrix Representation of Graphs

1.10 Matrix Representation of Graphs 42 Basic Concepts of Graphs 1.10 Matrix Representation of Graphs Definitions: In this section, we introduce two kinds of matrix representations of a graph, that is, the adjacency matrix and incidence matrix

More information

Iowa State University and American Institute of Mathematics

Iowa State University and American Institute of Mathematics Matrix and Matrix and Iowa State University and American Institute of Mathematics ICART 2008 May 28, 2008 Inverse Eigenvalue Problem for a Graph () Problem for a Graph of minimum rank Specific matrices

More information

Lecture 13: Spectral Graph Theory

Lecture 13: Spectral Graph Theory CSE 521: Design and Analysis of Algorithms I Winter 2017 Lecture 13: Spectral Graph Theory Lecturer: Shayan Oveis Gharan 11/14/18 Disclaimer: These notes have not been subjected to the usual scrutiny reserved

More information

Data Analysis and Manifold Learning Lecture 3: Graphs, Graph Matrices, and Graph Embeddings

Data Analysis and Manifold Learning Lecture 3: Graphs, Graph Matrices, and Graph Embeddings Data Analysis and Manifold Learning Lecture 3: Graphs, Graph Matrices, and Graph Embeddings Radu Horaud INRIA Grenoble Rhone-Alpes, France Radu.Horaud@inrialpes.fr http://perception.inrialpes.fr/ Outline

More information

Lecture 12: Introduction to Spectral Graph Theory, Cheeger s inequality

Lecture 12: Introduction to Spectral Graph Theory, Cheeger s inequality CSE 521: Design and Analysis of Algorithms I Spring 2016 Lecture 12: Introduction to Spectral Graph Theory, Cheeger s inequality Lecturer: Shayan Oveis Gharan May 4th Scribe: Gabriel Cadamuro Disclaimer:

More information

Spectral Graph Theory

Spectral Graph Theory Spectral raph Theory to appear in Handbook of Linear Algebra, second edition, CCR Press Steve Butler Fan Chung There are many different ways to associate a matrix with a graph (an introduction of which

More information

Laplacian Integral Graphs with Maximum Degree 3

Laplacian Integral Graphs with Maximum Degree 3 Laplacian Integral Graphs with Maximum Degree Steve Kirkland Department of Mathematics and Statistics University of Regina Regina, Saskatchewan, Canada S4S 0A kirkland@math.uregina.ca Submitted: Nov 5,

More information

Cospectral bipartite graphs for the normalized Laplacian. Steven Paul Osborne. A dissertation submitted to the graduate faculty

Cospectral bipartite graphs for the normalized Laplacian. Steven Paul Osborne. A dissertation submitted to the graduate faculty Cospectral bipartite graphs for the normalized Laplacian by Steven Paul Osborne A dissertation submitted to the graduate faculty in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY

More information

A lower bound for the Laplacian eigenvalues of a graph proof of a conjecture by Guo

A lower bound for the Laplacian eigenvalues of a graph proof of a conjecture by Guo A lower bound for the Laplacian eigenvalues of a graph proof of a conjecture by Guo A. E. Brouwer & W. H. Haemers 2008-02-28 Abstract We show that if µ j is the j-th largest Laplacian eigenvalue, and d

More information

18.312: Algebraic Combinatorics Lionel Levine. Lecture 19

18.312: Algebraic Combinatorics Lionel Levine. Lecture 19 832: Algebraic Combinatorics Lionel Levine Lecture date: April 2, 20 Lecture 9 Notes by: David Witmer Matrix-Tree Theorem Undirected Graphs Let G = (V, E) be a connected, undirected graph with n vertices,

More information

An Introduction to Algebraic Graph Theory

An Introduction to Algebraic Graph Theory An Introduction to Algebraic Graph Theory Rob Beezer beezer@ups.edu Department of Mathematics and Computer Science University of Puget Sound Mathematics Department Seminar Pacific University October 9,

More information

Markov Chains and Spectral Clustering

Markov Chains and Spectral Clustering Markov Chains and Spectral Clustering Ning Liu 1,2 and William J. Stewart 1,3 1 Department of Computer Science North Carolina State University, Raleigh, NC 27695-8206, USA. 2 nliu@ncsu.edu, 3 billy@ncsu.edu

More information

An Introduction to Spectral Graph Theory

An Introduction to Spectral Graph Theory An Introduction to Spectral Graph Theory Mackenzie Wheeler Supervisor: Dr. Gary MacGillivray University of Victoria WheelerM@uvic.ca Outline Outline 1. How many walks are there from vertices v i to v j

More information

Spectral Graph Theory

Spectral Graph Theory Spectral Graph Theory Aaron Mishtal April 27, 2016 1 / 36 Outline Overview Linear Algebra Primer History Theory Applications Open Problems Homework Problems References 2 / 36 Outline Overview Linear Algebra

More information

Linear algebra and applications to graphs Part 1

Linear algebra and applications to graphs Part 1 Linear algebra and applications to graphs Part 1 Written up by Mikhail Belkin and Moon Duchin Instructor: Laszlo Babai June 17, 2001 1 Basic Linear Algebra Exercise 1.1 Let V and W be linear subspaces

More information

Basic graph theory 18.S995 - L26.

Basic graph theory 18.S995 - L26. Basic graph theory 18.S995 - L26 dunkel@math.mit.edu http://java.dzone.com/articles/algorithm-week-graphs-and no cycles Isomorphic graphs f(a) = 1 f(b) = 6 f(c) = 8 f(d) = 3 f(g) = 5 f(h) = 2 f(i) = 4

More information

An Algorithmist s Toolkit September 10, Lecture 1

An Algorithmist s Toolkit September 10, Lecture 1 18.409 An Algorithmist s Toolkit September 10, 2009 Lecture 1 Lecturer: Jonathan Kelner Scribe: Jesse Geneson (2009) 1 Overview The class s goals, requirements, and policies were introduced, and topics

More information

Applied Mathematics Letters

Applied Mathematics Letters Applied Mathematics Letters (009) 15 130 Contents lists available at ScienceDirect Applied Mathematics Letters journal homepage: www.elsevier.com/locate/aml Spectral characterizations of sandglass graphs

More information

Networks and Their Spectra

Networks and Their Spectra Networks and Their Spectra Victor Amelkin University of California, Santa Barbara Department of Computer Science victor@cs.ucsb.edu December 4, 2017 1 / 18 Introduction Networks (= graphs) are everywhere.

More information

Eigenvalues of 2-edge-coverings

Eigenvalues of 2-edge-coverings Eigenvalues of -edge-coverings Steve Butler October 3, 007 Abstract A -edge-covering between G and H is an onto homomorphism from the vertices of G to the vertices of H so that each edge is covered twice

More information

Energy of Graphs. Sivaram K. Narayan Central Michigan University. Presented at CMU on October 10, 2015

Energy of Graphs. Sivaram K. Narayan Central Michigan University. Presented at CMU on October 10, 2015 Energy of Graphs Sivaram K. Narayan Central Michigan University Presented at CMU on October 10, 2015 1 / 32 Graphs We will consider simple graphs (no loops, no multiple edges). Let V = {v 1, v 2,..., v

More information

On graphs with largest Laplacian eigenvalue at most 4

On graphs with largest Laplacian eigenvalue at most 4 AUSTRALASIAN JOURNAL OF COMBINATORICS Volume 44 (2009), Pages 163 170 On graphs with largest Laplacian eigenvalue at most 4 G. R. Omidi Department of Mathematical Sciences Isfahan University of Technology

More information

Definition A finite Markov chain is a memoryless homogeneous discrete stochastic process with a finite number of states.

Definition A finite Markov chain is a memoryless homogeneous discrete stochastic process with a finite number of states. Chapter 8 Finite Markov Chains A discrete system is characterized by a set V of states and transitions between the states. V is referred to as the state space. We think of the transitions as occurring

More information

Spectral Generative Models for Graphs

Spectral Generative Models for Graphs Spectral Generative Models for Graphs David White and Richard C. Wilson Department of Computer Science University of York Heslington, York, UK wilson@cs.york.ac.uk Abstract Generative models are well known

More information

Program (Day 1) Spectral mapping theorem and asymptotic behavior of quantum walks on infinite graphs

Program (Day 1) Spectral mapping theorem and asymptotic behavior of quantum walks on infinite graphs Program (Day 1) Nov.29 (Fri) Kawai Hall, Dept. of Math., Tohoku Univ. 9:25 9:30 10:10 10:20 11:00 11:10 11:50 14:00 14:40 14:50 15:30 15:40 16:30 18:00 Opening Address from Chair Spectral mapping theorem

More information

Spectral characterization of Signed Graphs

Spectral characterization of Signed Graphs University of Primorska, Koper May 2015 Signed Graphs Basic notions on Signed Graphs Matrices of Signed Graphs A signed graph Γ is an ordered pair (G, σ), where G = (V (G), E(G)) is a graph and σ : E(G)

More information

The Simplest Construction of Expanders

The Simplest Construction of Expanders Spectral Graph Theory Lecture 14 The Simplest Construction of Expanders Daniel A. Spielman October 16, 2009 14.1 Overview I am going to present the simplest construction of expanders that I have been able

More information

Topics in Graph Theory

Topics in Graph Theory Topics in Graph Theory September 4, 2018 1 Preliminaries A graph is a system G = (V, E) consisting of a set V of vertices and a set E (disjoint from V ) of edges, together with an incidence function End

More information

Diffusion and random walks on graphs

Diffusion and random walks on graphs Diffusion and random walks on graphs Leonid E. Zhukov School of Data Analysis and Artificial Intelligence Department of Computer Science National Research University Higher School of Economics Structural

More information

Fundamentals of Matrices

Fundamentals of Matrices Maschinelles Lernen II Fundamentals of Matrices Christoph Sawade/Niels Landwehr/Blaine Nelson Tobias Scheffer Matrix Examples Recap: Data Linear Model: f i x = w i T x Let X = x x n be the data matrix

More information

THE NORMALIZED LAPLACIAN MATRIX AND

THE NORMALIZED LAPLACIAN MATRIX AND THE NORMALIZED LAPLACIAN MATRIX AND GENERAL RANDIĆ INDEX OF GRAPHS A Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Doctor

More information

Math 307 Learning Goals. March 23, 2010

Math 307 Learning Goals. March 23, 2010 Math 307 Learning Goals March 23, 2010 Course Description The course presents core concepts of linear algebra by focusing on applications in Science and Engineering. Examples of applications from recent

More information

Chapter 2 Spectra of Finite Graphs

Chapter 2 Spectra of Finite Graphs Chapter 2 Spectra of Finite Graphs 2.1 Characteristic Polynomials Let G = (V, E) be a finite graph on n = V vertices. Numbering the vertices, we write down its adjacency matrix in an explicit form of n

More information

Two Laplacians for the distance matrix of a graph

Two Laplacians for the distance matrix of a graph Two Laplacians for the distance matrix of a graph Mustapha Aouchiche and Pierre Hansen GERAD and HEC Montreal, Canada CanaDAM 2013, Memorial University, June 1013 Aouchiche & Hansen CanaDAM 2013, Memorial

More information

ELA A LOWER BOUND FOR THE SECOND LARGEST LAPLACIAN EIGENVALUE OF WEIGHTED GRAPHS

ELA A LOWER BOUND FOR THE SECOND LARGEST LAPLACIAN EIGENVALUE OF WEIGHTED GRAPHS A LOWER BOUND FOR THE SECOND LARGEST LAPLACIAN EIGENVALUE OF WEIGHTED GRAPHS ABRAHAM BERMAN AND MIRIAM FARBER Abstract. Let G be a weighted graph on n vertices. Let λ n 1 (G) be the second largest eigenvalue

More information

Spectra of Adjacency and Laplacian Matrices

Spectra of Adjacency and Laplacian Matrices Spectra of Adjacency and Laplacian Matrices Definition: University of Alicante (Spain) Matrix Computing (subject 3168 Degree in Maths) 30 hours (theory)) + 15 hours (practical assignment) Contents 1. Spectra

More information

1 T 1 = where 1 is the all-ones vector. For the upper bound, let v 1 be the eigenvector corresponding. u:(u,v) E v 1(u)

1 T 1 = where 1 is the all-ones vector. For the upper bound, let v 1 be the eigenvector corresponding. u:(u,v) E v 1(u) CME 305: Discrete Mathematics and Algorithms Instructor: Reza Zadeh (rezab@stanford.edu) Final Review Session 03/20/17 1. Let G = (V, E) be an unweighted, undirected graph. Let λ 1 be the maximum eigenvalue

More information

Laplacians of Graphs, Spectra and Laplacian polynomials

Laplacians of Graphs, Spectra and Laplacian polynomials Laplacians of Graphs, Spectra and Laplacian polynomials Lector: Alexander Mednykh Sobolev Institute of Mathematics Novosibirsk State University Winter School in Harmonic Functions on Graphs and Combinatorial

More information

On marker set distance Laplacian eigenvalues in graphs.

On marker set distance Laplacian eigenvalues in graphs. Malaya Journal of Matematik, Vol 6, No 2, 369-374, 2018 https://doiorg/1026637/mjm0602/0011 On marker set distance Laplacian eigenvalues in graphs Medha Itagi Huilgol1 * and S Anuradha2 Abstract In our

More information

Constructions for cospectral graphs for the normalized Laplacian matrix and. distance matrix. Kristin Heysse

Constructions for cospectral graphs for the normalized Laplacian matrix and. distance matrix. Kristin Heysse Constructions for cospectral graphs for the normalized Laplacian matrix and distance matrix by Kristin Heysse A dissertation submitted to the graduate faculty in partial fulfillment of the requirements

More information

Algebraic Representation of Networks

Algebraic Representation of Networks Algebraic Representation of Networks 0 1 2 1 1 0 0 1 2 0 0 1 1 1 1 1 Hiroki Sayama sayama@binghamton.edu Describing networks with matrices (1) Adjacency matrix A matrix with rows and columns labeled by

More information

The least eigenvalue of the signless Laplacian of non-bipartite unicyclic graphs with k pendant vertices

The least eigenvalue of the signless Laplacian of non-bipartite unicyclic graphs with k pendant vertices Electronic Journal of Linear Algebra Volume 26 Volume 26 (2013) Article 22 2013 The least eigenvalue of the signless Laplacian of non-bipartite unicyclic graphs with k pendant vertices Ruifang Liu rfliu@zzu.edu.cn

More information

1 Matrix notation and preliminaries from spectral graph theory

1 Matrix notation and preliminaries from spectral graph theory Graph clustering (or community detection or graph partitioning) is one of the most studied problems in network analysis. One reason for this is that there are a variety of ways to define a cluster or community.

More information

Learning from Sensor Data: Set II. Behnaam Aazhang J.S. Abercombie Professor Electrical and Computer Engineering Rice University

Learning from Sensor Data: Set II. Behnaam Aazhang J.S. Abercombie Professor Electrical and Computer Engineering Rice University Learning from Sensor Data: Set II Behnaam Aazhang J.S. Abercombie Professor Electrical and Computer Engineering Rice University 1 6. Data Representation The approach for learning from data Probabilistic

More information

A linear algebraic view of partition regular matrices

A linear algebraic view of partition regular matrices A linear algebraic view of partition regular matrices Leslie Hogben Jillian McLeod June 7, 3 4 5 6 7 8 9 Abstract Rado showed that a rational matrix is partition regular over N if and only if it satisfies

More information

Markov Chains, Random Walks on Graphs, and the Laplacian

Markov Chains, Random Walks on Graphs, and the Laplacian Markov Chains, Random Walks on Graphs, and the Laplacian CMPSCI 791BB: Advanced ML Sridhar Mahadevan Random Walks! There is significant interest in the problem of random walks! Markov chain analysis! Computer

More information

Math 443/543 Graph Theory Notes 5: Graphs as matrices, spectral graph theory, and PageRank

Math 443/543 Graph Theory Notes 5: Graphs as matrices, spectral graph theory, and PageRank Math 443/543 Graph Theory Notes 5: Graphs as matrices, spectral graph theory, and PageRank David Glickenstein November 3, 4 Representing graphs as matrices It will sometimes be useful to represent graphs

More information

Preliminaries and Complexity Theory

Preliminaries and Complexity Theory Preliminaries and Complexity Theory Oleksandr Romanko CAS 746 - Advanced Topics in Combinatorial Optimization McMaster University, January 16, 2006 Introduction Book structure: 2 Part I Linear Algebra

More information

Matching Polynomials of Graphs

Matching Polynomials of Graphs Spectral Graph Theory Lecture 25 Matching Polynomials of Graphs Daniel A Spielman December 7, 2015 Disclaimer These notes are not necessarily an accurate representation of what happened in class The notes

More information

The non-bipartite graphs with all but two eigenvalues in

The non-bipartite graphs with all but two eigenvalues in The non-bipartite graphs with all but two eigenvalues in [ 1, 1] L.S. de Lima 1, A. Mohammadian 1, C.S. Oliveira 2 1 Departamento de Engenharia de Produção, Centro Federal de Educação Tecnológica Celso

More information

Inverse Perron values and connectivity of a uniform hypergraph

Inverse Perron values and connectivity of a uniform hypergraph Inverse Perron values and connectivity of a uniform hypergraph Changjiang Bu College of Automation College of Science Harbin Engineering University Harbin, PR China buchangjiang@hrbeu.edu.cn Jiang Zhou

More information

Data Analysis and Manifold Learning Lecture 7: Spectral Clustering

Data Analysis and Manifold Learning Lecture 7: Spectral Clustering Data Analysis and Manifold Learning Lecture 7: Spectral Clustering Radu Horaud INRIA Grenoble Rhone-Alpes, France Radu.Horaud@inrialpes.fr http://perception.inrialpes.fr/ Outline of Lecture 7 What is spectral

More information

17.1 Directed Graphs, Undirected Graphs, Incidence Matrices, Adjacency Matrices, Weighted Graphs

17.1 Directed Graphs, Undirected Graphs, Incidence Matrices, Adjacency Matrices, Weighted Graphs Chapter 17 Graphs and Graph Laplacians 17.1 Directed Graphs, Undirected Graphs, Incidence Matrices, Adjacency Matrices, Weighted Graphs Definition 17.1. A directed graph isa pairg =(V,E), where V = {v

More information

Eigenvalues, random walks and Ramanujan graphs

Eigenvalues, random walks and Ramanujan graphs Eigenvalues, random walks and Ramanujan graphs David Ellis 1 The Expander Mixing lemma We have seen that a bounded-degree graph is a good edge-expander if and only if if has large spectral gap If G = (V,

More information

Graph fundamentals. Matrices associated with a graph

Graph fundamentals. Matrices associated with a graph Graph fundamentals Matrices associated with a graph Drawing a picture of a graph is one way to represent it. Another type of representation is via a matrix. Let G be a graph with V (G) ={v 1,v,...,v n

More information

22m:033 Notes: 7.1 Diagonalization of Symmetric Matrices

22m:033 Notes: 7.1 Diagonalization of Symmetric Matrices m:33 Notes: 7. Diagonalization of Symmetric Matrices Dennis Roseman University of Iowa Iowa City, IA http://www.math.uiowa.edu/ roseman May 3, Symmetric matrices Definition. A symmetric matrix is a matrix

More information

Lecture 12 : Graph Laplacians and Cheeger s Inequality

Lecture 12 : Graph Laplacians and Cheeger s Inequality CPS290: Algorithmic Foundations of Data Science March 7, 2017 Lecture 12 : Graph Laplacians and Cheeger s Inequality Lecturer: Kamesh Munagala Scribe: Kamesh Munagala Graph Laplacian Maybe the most beautiful

More information

Laplacian and Random Walks on Graphs

Laplacian and Random Walks on Graphs Laplacian and Random Walks on Graphs Linyuan Lu University of South Carolina Selected Topics on Spectral Graph Theory (II) Nankai University, Tianjin, May 22, 2014 Five talks Selected Topics on Spectral

More information

Lecture 1: Graphs, Adjacency Matrices, Graph Laplacian

Lecture 1: Graphs, Adjacency Matrices, Graph Laplacian Lecture 1: Graphs, Adjacency Matrices, Graph Laplacian Radu Balan January 31, 2017 G = (V, E) An undirected graph G is given by two pieces of information: a set of vertices V and a set of edges E, G =

More information

Fiedler s Theorems on Nodal Domains

Fiedler s Theorems on Nodal Domains Spectral Graph Theory Lecture 7 Fiedler s Theorems on Nodal Domains Daniel A Spielman September 9, 202 7 About these notes These notes are not necessarily an accurate representation of what happened in

More information

Study Notes on Matrices & Determinants for GATE 2017

Study Notes on Matrices & Determinants for GATE 2017 Study Notes on Matrices & Determinants for GATE 2017 Matrices and Determinates are undoubtedly one of the most scoring and high yielding topics in GATE. At least 3-4 questions are always anticipated from

More information

The Matrix-Tree Theorem

The Matrix-Tree Theorem The Matrix-Tree Theorem Christopher Eur March 22, 2015 Abstract: We give a brief introduction to graph theory in light of linear algebra. Our results culminates in the proof of Matrix-Tree Theorem. 1 Preliminaries

More information

Section 4.5 Eigenvalues of Symmetric Tridiagonal Matrices

Section 4.5 Eigenvalues of Symmetric Tridiagonal Matrices Section 4.5 Eigenvalues of Symmetric Tridiagonal Matrices Key Terms Symmetric matrix Tridiagonal matrix Orthogonal matrix QR-factorization Rotation matrices (plane rotations) Eigenvalues We will now complete

More information

CS168: The Modern Algorithmic Toolbox Lectures #11 and #12: Spectral Graph Theory

CS168: The Modern Algorithmic Toolbox Lectures #11 and #12: Spectral Graph Theory CS168: The Modern Algorithmic Toolbox Lectures #11 and #12: Spectral Graph Theory Tim Roughgarden & Gregory Valiant May 2, 2016 Spectral graph theory is the powerful and beautiful theory that arises from

More information

Computability and Complexity Theory: An Introduction

Computability and Complexity Theory: An Introduction Computability and Complexity Theory: An Introduction meena@imsc.res.in http://www.imsc.res.in/ meena IMI-IISc, 20 July 2006 p. 1 Understanding Computation Kinds of questions we seek answers to: Is a given

More information

Ma/CS 6b Class 23: Eigenvalues in Regular Graphs

Ma/CS 6b Class 23: Eigenvalues in Regular Graphs Ma/CS 6b Class 3: Eigenvalues in Regular Graphs By Adam Sheffer Recall: The Spectrum of a Graph Consider a graph G = V, E and let A be the adjacency matrix of G. The eigenvalues of G are the eigenvalues

More information

The graph isomorphism problem and the structure of walks

The graph isomorphism problem and the structure of walks The graph isomorphism problem and the structure of walks Péter Madarasi Applied Mathematics MSc Thesis Supervisor: Alpár Jüttner senior research fellow ELTE Institute of Mathematics, Department of Operations

More information

Spectral Graph Theory and its Applications. Daniel A. Spielman Dept. of Computer Science Program in Applied Mathematics Yale Unviersity

Spectral Graph Theory and its Applications. Daniel A. Spielman Dept. of Computer Science Program in Applied Mathematics Yale Unviersity Spectral Graph Theory and its Applications Daniel A. Spielman Dept. of Computer Science Program in Applied Mathematics Yale Unviersity Outline Adjacency matrix and Laplacian Intuition, spectral graph drawing

More information

Spanning Trees of Shifted Simplicial Complexes

Spanning Trees of Shifted Simplicial Complexes Art Duval (University of Texas at El Paso) Caroline Klivans (Brown University) Jeremy Martin (University of Kansas) Special Session on Extremal and Probabilistic Combinatorics University of Nebraska, Lincoln

More information

Spectral Theory of Unsigned and Signed Graphs Applications to Graph Clustering. Some Slides

Spectral Theory of Unsigned and Signed Graphs Applications to Graph Clustering. Some Slides Spectral Theory of Unsigned and Signed Graphs Applications to Graph Clustering Some Slides Jean Gallier Department of Computer and Information Science University of Pennsylvania Philadelphia, PA 19104,

More information

MATH 829: Introduction to Data Mining and Analysis Clustering II

MATH 829: Introduction to Data Mining and Analysis Clustering II his lecture is based on U. von Luxburg, A Tutorial on Spectral Clustering, Statistics and Computing, 17 (4), 2007. MATH 829: Introduction to Data Mining and Analysis Clustering II Dominique Guillot Departments

More information

Chapter 3 Transformations

Chapter 3 Transformations Chapter 3 Transformations An Introduction to Optimization Spring, 2014 Wei-Ta Chu 1 Linear Transformations A function is called a linear transformation if 1. for every and 2. for every If we fix the bases

More information

Spectral Clustering. Guokun Lai 2016/10

Spectral Clustering. Guokun Lai 2016/10 Spectral Clustering Guokun Lai 2016/10 1 / 37 Organization Graph Cut Fundamental Limitations of Spectral Clustering Ng 2002 paper (if we have time) 2 / 37 Notation We define a undirected weighted graph

More information

Lecture 1 and 2: Introduction and Graph theory basics. Spring EE 194, Networked estimation and control (Prof. Khan) January 23, 2012

Lecture 1 and 2: Introduction and Graph theory basics. Spring EE 194, Networked estimation and control (Prof. Khan) January 23, 2012 Lecture 1 and 2: Introduction and Graph theory basics Spring 2012 - EE 194, Networked estimation and control (Prof. Khan) January 23, 2012 Spring 2012: EE-194-02 Networked estimation and control Schedule

More information

Data Analysis and Manifold Learning Lecture 9: Diffusion on Manifolds and on Graphs

Data Analysis and Manifold Learning Lecture 9: Diffusion on Manifolds and on Graphs Data Analysis and Manifold Learning Lecture 9: Diffusion on Manifolds and on Graphs Radu Horaud INRIA Grenoble Rhone-Alpes, France Radu.Horaud@inrialpes.fr http://perception.inrialpes.fr/ Outline of Lecture

More information

1 Adjacency matrix and eigenvalues

1 Adjacency matrix and eigenvalues CSC 5170: Theory of Computational Complexity Lecture 7 The Chinese University of Hong Kong 1 March 2010 Our objective of study today is the random walk algorithm for deciding if two vertices in an undirected

More information

Lecture 1. 1 if (i, j) E a i,j = 0 otherwise, l i,j = d i if i = j, and

Lecture 1. 1 if (i, j) E a i,j = 0 otherwise, l i,j = d i if i = j, and Specral Graph Theory and its Applications September 2, 24 Lecturer: Daniel A. Spielman Lecture. A quick introduction First of all, please call me Dan. If such informality makes you uncomfortable, you can

More information

ORIE 6334 Spectral Graph Theory September 22, Lecture 11

ORIE 6334 Spectral Graph Theory September 22, Lecture 11 ORIE 6334 Spectral Graph Theory September, 06 Lecturer: David P. Williamson Lecture Scribe: Pu Yang In today s lecture we will focus on discrete time random walks on undirected graphs. Specifically, we

More information

Average Mixing Matrix of Trees

Average Mixing Matrix of Trees Electronic Journal of Linear Algebra Volume 34 Volume 34 08 Article 9 08 Average Mixing Matrix of Trees Chris Godsil University of Waterloo, cgodsil@uwaterloo.ca Krystal Guo Université libre de Bruxelles,

More information

SPECTRUM OF (k, r) - REGULAR HYPERGRAPH

SPECTRUM OF (k, r) - REGULAR HYPERGRAPH SPECTRUM OF (k, r) - REGULAR HYPERGRAPH K Reji Kumar 1 and Renny P Varghese Abstract. We present a spectral theory of uniform, regular and linear hypergraph. The main result are the nature of the eigen

More information

On the distance signless Laplacian spectral radius of graphs and digraphs

On the distance signless Laplacian spectral radius of graphs and digraphs Electronic Journal of Linear Algebra Volume 3 Volume 3 (017) Article 3 017 On the distance signless Laplacian spectral radius of graphs and digraphs Dan Li Xinjiang University,Urumqi, ldxjedu@163.com Guoping

More information

Normalized Laplacian spectrum of two new types of join graphs

Normalized Laplacian spectrum of two new types of join graphs Journal of Linear and Topological Algebra Vol. 6, No. 1, 217, 1-9 Normalized Laplacian spectrum of two new types of join graphs M. Hakimi-Nezhaad a, M. Ghorbani a a Department of Mathematics, Faculty of

More information

Notes on Linear Algebra and Matrix Theory

Notes on Linear Algebra and Matrix Theory Massimo Franceschet featuring Enrico Bozzo Scalar product The scalar product (a.k.a. dot product or inner product) of two real vectors x = (x 1,..., x n ) and y = (y 1,..., y n ) is not a vector but a

More information

Semidefinite and Second Order Cone Programming Seminar Fall 2001 Lecture 5

Semidefinite and Second Order Cone Programming Seminar Fall 2001 Lecture 5 Semidefinite and Second Order Cone Programming Seminar Fall 2001 Lecture 5 Instructor: Farid Alizadeh Scribe: Anton Riabov 10/08/2001 1 Overview We continue studying the maximum eigenvalue SDP, and generalize

More information

Algebraic aspects of the normalized Laplacian

Algebraic aspects of the normalized Laplacian Algebraic aspects of the normalized Laplacian Steve Butler Abstract Spectral graph theory looks at the interplay between the structure of a graph and the eigenvalues of a matrix associated with the graph.

More information

ONE CAN HEAR THE EULER CHARACTERISTIC OF A SIMPLICIAL COMPLEX

ONE CAN HEAR THE EULER CHARACTERISTIC OF A SIMPLICIAL COMPLEX ONE CAN HEAR THE EULER CHARACTERISTIC OF A SIMPLICIAL COMPLEX OLIVER KNILL y G g(x, y) with g = L 1 but also agrees with Abstract. We prove that that the number p of positive eigenvalues of the connection

More information

Eigenvalues and Eigenvectors

Eigenvalues and Eigenvectors Eigenvalues and Eigenvectors Philippe B. Laval KSU Fall 2015 Philippe B. Laval (KSU) Eigenvalues and Eigenvectors Fall 2015 1 / 14 Introduction We define eigenvalues and eigenvectors. We discuss how to

More information

Math 307 Learning Goals

Math 307 Learning Goals Math 307 Learning Goals May 14, 2018 Chapter 1 Linear Equations 1.1 Solving Linear Equations Write a system of linear equations using matrix notation. Use Gaussian elimination to bring a system of linear

More information

series. Utilize the methods of calculus to solve applied problems that require computational or algebraic techniques..

series. Utilize the methods of calculus to solve applied problems that require computational or algebraic techniques.. 1 Use computational techniques and algebraic skills essential for success in an academic, personal, or workplace setting. (Computational and Algebraic Skills) MAT 203 MAT 204 MAT 205 MAT 206 Calculus I

More information

Spectral Algorithms I. Slides based on Spectral Mesh Processing Siggraph 2010 course

Spectral Algorithms I. Slides based on Spectral Mesh Processing Siggraph 2010 course Spectral Algorithms I Slides based on Spectral Mesh Processing Siggraph 2010 course Why Spectral? A different way to look at functions on a domain Why Spectral? Better representations lead to simpler solutions

More information

Lecture 10. Lecturer: Aleksander Mądry Scribes: Mani Bastani Parizi and Christos Kalaitzis

Lecture 10. Lecturer: Aleksander Mądry Scribes: Mani Bastani Parizi and Christos Kalaitzis CS-621 Theory Gems October 18, 2012 Lecture 10 Lecturer: Aleksander Mądry Scribes: Mani Bastani Parizi and Christos Kalaitzis 1 Introduction In this lecture, we will see how one can use random walks to

More information

Graph Sparsification III: Ramanujan Graphs, Lifts, and Interlacing Families

Graph Sparsification III: Ramanujan Graphs, Lifts, and Interlacing Families Graph Sparsification III: Ramanujan Graphs, Lifts, and Interlacing Families Nikhil Srivastava Microsoft Research India Simons Institute, August 27, 2014 The Last Two Lectures Lecture 1. Every weighted

More information

Lecture 13 Spectral Graph Algorithms

Lecture 13 Spectral Graph Algorithms COMS 995-3: Advanced Algorithms March 6, 7 Lecture 3 Spectral Graph Algorithms Instructor: Alex Andoni Scribe: Srikar Varadaraj Introduction Today s topics: Finish proof from last lecture Example of random

More information

FORBIDDEN MINORS FOR THE CLASS OF GRAPHS G WITH ξ(g) 2. July 25, 2006

FORBIDDEN MINORS FOR THE CLASS OF GRAPHS G WITH ξ(g) 2. July 25, 2006 FORBIDDEN MINORS FOR THE CLASS OF GRAPHS G WITH ξ(g) 2 LESLIE HOGBEN AND HEIN VAN DER HOLST July 25, 2006 Abstract. For a given simple graph G, S(G) is defined to be the set of real symmetric matrices

More information

On the distance and distance signless Laplacian eigenvalues of graphs and the smallest Gersgorin disc

On the distance and distance signless Laplacian eigenvalues of graphs and the smallest Gersgorin disc Electronic Journal of Linear Algebra Volume 34 Volume 34 (018) Article 14 018 On the distance and distance signless Laplacian eigenvalues of graphs and the smallest Gersgorin disc Fouzul Atik Indian Institute

More information

Physical Metaphors for Graphs

Physical Metaphors for Graphs Graphs and Networks Lecture 3 Physical Metaphors for Graphs Daniel A. Spielman October 9, 203 3. Overview We will examine physical metaphors for graphs. We begin with a spring model and then discuss resistor

More information

Very few Moore Graphs

Very few Moore Graphs Very few Moore Graphs Anurag Bishnoi June 7, 0 Abstract We prove here a well known result in graph theory, originally proved by Hoffman and Singleton, that any non-trivial Moore graph of diameter is regular

More information

Spectral Graph Theory and You: Matrix Tree Theorem and Centrality Metrics

Spectral Graph Theory and You: Matrix Tree Theorem and Centrality Metrics Spectral Graph Theory and You: and Centrality Metrics Jonathan Gootenberg March 11, 2013 1 / 19 Outline of Topics 1 Motivation Basics of Spectral Graph Theory Understanding the characteristic polynomial

More information