On Hadamard s Maximal Determinant Problem
|
|
- Branden Hines
- 5 years ago
- Views:
Transcription
1 Judy-anne Osborn MSI, ANU April 2009
2 m m 1 max det =?
3 A Naive Computer Search Order max det Time 1 1 fast 2 1 fast 3 2 fast 4 3 fast 5 5 fast 6 9 order of days 7 32 order of years 8 56 order of the age of the Universe
4 As well by hand? I found nested Max Dets }2}3}5}9 }1} 1
5 As well by hand? I found nested Max Dets }2}3}5}9 }1} 1
6 As well by hand? I found nested Max Dets }2}3}5}9 }1} 1
7 As well by hand? I found nested Max Dets }2}3}5}9 }1} 1
8 As well by hand? I found nested Max Dets }2}3}5}9 }1} 1 But no further!
9 The problem turns out to be famous Hadamard s Maximal Determinant Problem was posed in 1893 Jacques Hadamard
10 A little selected history on this century-old question...
11 Observe: 2D: ( ) 0 1 or ( ) 1 1 ( ) 1 0 ( ) 1 0 ( 0 ) 1 ( 1 )
12 Geometry: max det = max (hyper-)volume 3D:
13 An equivalent problem: {+1, 1} matrices ( ( ( ( m m matrix ( ( ( ( ( 2) border add row 1 ( ( column ops ( row ops (
14 An equivalent problem: {+1, 1} matrices ( ( ( ( m m matrix ( ( ( ( ( 2) border add row 1 det new = 2 m det old ( ( column ops ( row ops (
15 Volume interpretation upper bound on max det ±1 ±1 max.... }. n ±1 ±1 } n What is the upper bound?
16 Volume interpretation upper bound on max det ±1 ±1 max.... }. n ±1 ±1 } n What is the upper bound? ( (±1) (±1) 2 ) n = n n/2
17 Volume interpretation upper bound on max det ±1 ±1 max.... }. n ±1 ±1 } n What is the upper bound? Why? ( (±1) (±1) 2 ) n = n n/2
18 Volume interpretation upper bound on max det ±1 ±1 max.... }. n ±1 ±1 } n What is the upper bound? ( (±1) (±1) 2 ) n = n n/2 Why? (Columns/rows orthogonal)
19 When is the bound tight? Tight when {+1, 1} square matrix H of order n satisfies HH T = ni
20 When is the bound tight? Tight when {+1, 1} square matrix H of order n satisfies HH T = ni H is called a Hadamard Matrix.
21 When is the bound tight? Tight when {+1, 1} square matrix H of order n satisfies HH T = ni H is called a Hadamard Matrix. A necessary condition on existence of H is: n = 1, 2 or n 0 (mod 4)
22 When is the bound tight? Tight when {+1, 1} square matrix H of order n satisfies HH T = ni H is called a Hadamard Matrix. A necessary condition on existence of H is: n = 1, 2 or n 0 (mod 4) Hadamard Conjecture (Paley, 1933): this is also sufficient.
23 Evidence for Hadamard Conjecture Many constructions for infinite families, including Sylvester, 2 r First Paley, using finite fields, p r + 1, p prime Second Paley, using finite fields, 2p r + 2, p prime
24 Evidence for Hadamard Conjecture Many constructions for infinite families, including Sylvester, 2 r First Paley, using finite fields, p r + 1, p prime Second Paley, using finite fields, 2p r + 2, p prime Other constructions and ad hoc examples due to people including Williamson Jenny Seberry
25 Evidence for Hadamard Conjecture Many constructions for infinite families, including Sylvester, 2 r First Paley, using finite fields, p r + 1, p prime Second Paley, using finite fields, 2p r + 2, p prime Other constructions and ad hoc examples due to people including Williamson Jenny Seberry Smallest n 0 (mod 4) currently undecided: n = 668.
26 More evidence Order Number of inequivalent Hadamard matrices see Sloan s sequence A
27 Max Dets for non-hadamard orders?
28 Max Dets for non-hadamard orders? n max det 2 n n max det 2 n n max det 2 n
29 Max Dets for non-hadamard orders? n max det 2 n n max det 2 n n max det 2 n
30 Max Dets for non-hadamard orders? n max det 2 n n max det 2 n n max det 2 n
31 Tighter upper bounds? The Hadamard bound of n n/2 holds for all orders but is never tight for n 0 (mod 4) when n > 2.
32 Tighter upper bounds? The Hadamard bound of n n/2 holds for all orders but is never tight for n 0 (mod 4) when n > 2. Better upper bounds for non-hadamard orders were proved by Barba in 1933 Ehlich in 1962, 64 Wojtas in 1964 Cohn (proved a new bound tightness result) in 2000
33 The best known upper bounds are: The Barba-Ehlich bound holds for n 1 (mod 4): 2n 1(n 1) (n 1)/2
34 The best known upper bounds are: The Barba-Ehlich bound holds for n 1 (mod 4): 2n 1(n 1) (n 1)/2 The Ehlich-Wojtas bound holds for n 2 (mod 4): (2n 2)(n 2) (n/2) 1
35 The best known upper bounds are: The Ehlich bound holds for n 3 (mod 4): (n 3) n s u v ur 2 (n 3+4r) 2 (n+1+4r) 2 1 n 3 + 4r v(r + 1) n r where s = 3 for n = 3, s = 5 for n = 7, s = 5 or 6 for n = 11, s = 6 for n = 15, 19,..., 59, and s = 7 for n 63, r = n s, n = rs + v and u = s v.
36 Percentages of bounds met: summary from Will Orrick s: maxdet
37 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n.
38 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n. Consider gram matrix G := RR T
39 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n. Consider gram matrix G := RR T 1. G has all n s on the diagonal
40 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n. Consider gram matrix G := RR T 1. G has all n s on the diagonal 2. G is positive definite off-diagonal entries have size < n
41 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n. Consider gram matrix G := RR T 1. G has all n s on the diagonal 2. G is positive definite off-diagonal entries have size < n 3. G has all off-diagonal entries n (mod 2)
42 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n. Consider gram matrix G := RR T 1. G has all n s on the diagonal 2. G is positive definite off-diagonal entries have size < n 3. G has all off-diagonal entries n (mod 2) 4. G has all off-diagonal entries n (mod 4) with R normalized
43 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n. Consider gram matrix G := RR T 1. G has all n s on the diagonal 2. G is positive definite off-diagonal entries have size < n 3. G has all off-diagonal entries n (mod 2) 4. G has all off-diagonal entries n (mod 4) with R normalized 5. G is symmetric
44 The idea behind the bounds Let R be a maximal determinant square ±1 matrix of order n. Consider gram matrix G := RR T 1. G has all n s on the diagonal 2. G is positive definite off-diagonal entries have size < n 3. G has all off-diagonal entries n (mod 2) 4. G has all off-diagonal entries n (mod 4) with R normalized 5. G is symmetric Let G n be the set of all gram matrices, G, of order n
45 The idea behind the bounds, continued Let G n be the set of matrices for which properties 1 5 hold.
46 The idea behind the bounds, continued Let G n be the set of matrices for which properties 1 5 hold. Then G n G n
47 The idea behind the bounds, continued Let G n be the set of matrices for which properties 1 5 hold. Then G n G n Hence max det ( Gn ) max det (G n )
48 Case: n 1 (mod 4) The matrix which was proven by Barba and Ehlich to have largest determinant in G n is n n
49 Case: n 2 (mod 4) The matrix which was proven by Ehlich and Wojtas to have largest determinant in G n is ( ) n 2 F 0, where F =... 0 F 2 n
50 Case: n 3 (mod 4) We expect a matrix with largest determinant in G n to be: n n In general, this is wrong Ehlich proved: the correct best determinant matrix in G n is a block form with off-diagonal entries from the set { 1, +3}.
51 Structure in the n 1 (mod 4) case What is a necessary condition for tightness of: Barba-Ehlich: 2n 1(n 1) (n 1)/2?
52 Structure in the n 1 (mod 4) case What is a necessary condition for tightness of: Barba-Ehlich: 2n 1(n 1) (n 1)/2? Lemma: The number 2n 1 is a perfect square iff q N such that n = q 2 + (q + 1) 2
53 In the literature, a conjecture on tightness: Conjecture: The Barba-Ehlich bound is tight whenever n is a sum of two consecutive squares: n = q 2 + (q + 1) 2
54 In the literature, a conjecture on tightness: Conjecture: The Barba-Ehlich bound is tight whenever n is a sum of two consecutive squares: n = q 2 + (q + 1) 2 Evidence: True for q = 2, 4
55 In the literature, a conjecture on tightness: Conjecture: The Barba-Ehlich bound is tight whenever n is a sum of two consecutive squares: n = q 2 + (q + 1) 2 Evidence: True for q = 2, 4 and when q = p r for p an odd prime proved by Brouwer s Construction.
56 What chance an exact maxdet formula n 1 (mod 4)? n max det 2 n Guess
57 What chance an exact maxdet formula n 1 (mod 4)? n max det 2 n Guess A guess/conjecture due to Will Orrick is that n = 4k + 1 is always divisible by max det 2 n 1 for k 2k 1,
58 What chance an exact maxdet formula n 1 (mod 4)? n max det 2 n Guess A guess/conjecture due to Will Orrick is that n = 4k + 1 is always divisible by max det 2 n 1 for k 2k 1, with coefficients growing quadratically between n s for which n = q 2 + (q + 1) 2.
59 What can we hope to compute? The ideas of Ehlich and Wojtas were reused by Chadjipantelis, Kounias and Moissiadis in the 1980 s to find max det matrices for n = 17 and n = 21.
60 What can we hope to compute? The ideas of Ehlich and Wojtas were reused by Chadjipantelis, Kounias and Moissiadis in the 1980 s to find max det matrices for n = 17 and n = 21. Will Orrick used similar ideas in the 2000 s to prove maximal an n = 15 matrix that had previously been found by Cohn; as well as filling in some gaps in CKM s published proofs.
61 What can we hope to compute? The ideas of Ehlich and Wojtas were reused by Chadjipantelis, Kounias and Moissiadis in the 1980 s to find max det matrices for n = 17 and n = 21. Will Orrick used similar ideas in the 2000 s to prove maximal an n = 15 matrix that had previously been found by Cohn; as well as filling in some gaps in CKM s published proofs. Are we within reach of n = 29?
62 Basic Idea Two steps: 1. Find candidate gram matrices in G n. 2. Check if candidates decompose in the form RR T.
63 Essentials of Step 1. There exist theorems which bound the determinants of candidate gram matrices in terms of their sub-matrices. So we can set a target determinant and build candidates: pruning too-small sub-matrices as we go. n n n n n
64 Computational considerations for Step 1. We must have efficient ways to calculate determinants! Rank-One Update Theorems: O(size 3 ) O(size 2 ), at the expense of some book-keeping.
65 Computational considerations for Step 1. ( Need to prune equivalent gram matrices: eg. ( ( ( under simultaneous row and column permutation
66 Computational considerations for Step 1. So need a (partial ordering) which prunes heavily enough, and is practical to compute on-the-fly?
67 Computational considerations for Step 1. So need a (partial ordering) which prunes heavily enough, and is practical to compute on-the-fly? Make sure we don t miss any valid candidates!
68 Essentials of Step 2. Consider gram candidates A and B whose determinants agree.
69 Essentials of Step 2. Consider gram candidates A and B whose determinants agree. These are candidates for RR T and R T R
70 Essentials of Step 2. Consider gram candidates A and B whose determinants agree. These are candidates for RR T and R T R Implement two kinds of constraints: Linear, Quadratic which use A and B simultaneously
71 Essentials of Step 2. the linear constraints Let A = (a ij ) and B = (b ij ).
72 Essentials of Step 2. the linear constraints Let A = (a ij ) and B = (b ij ). Assume r 0 R =. r n 1 = ( ) c 0 c n 1
73 Essentials of Step 2. the linear constraints Let A = (a ij ) and B = (b ij ). Assume r 0 R =. r n 1 = ( ) c 0 c n 1 Then a ij = r i.r j and b ij = c i.c j
74 Essentials of Step 2. the linear constraints Let A = (a ij ) and B = (b ij ). Assume r 0 R =. r n 1 = ( ) c 0 c n 1 Then a ij = r i.r j and b ij = c i.c j We need to implement an ordering to prune duplicates
75 Considerations for the linear constraints of Step 2. eg A = If we have already built r 0 = (1,,, 1, 1,,,,,, 1, 1, 1, 1, 1, 1, 1) r 1 = (,, 1,,,,, 1, 1, 1,,,, 1, 1, 1, 1) then r 2 breaks into blocks: r 2 = (a; b; c; d, e; f, g; h, i, j; k, l, m; n, o, p, q) and
76 Considerations for the linear constraints of Step 2. eg A = If we have already built r 0 = (1,,, 1, 1,,,,,, 1, 1, 1, 1, 1, 1, 1) r 1 = (,, 1,,,,, 1, 1, 1,,,, 1, 1, 1, 1) then r 2 breaks into blocks: r 2 = (a; b; c; d, e; f, g; h, i, j; k, l, m; n, o, p, q) and Adding more rows is a process of successive block refinement
77 Considerations for the linear constraints of Step 2. Because we work with both rows and columns, we need a way of refining row-blocks and column blocks simultaneously! ?
78 Essentials of Step 2. the quadratic constraints To derive the quadratic constraints, write key ingredients in block form: ( ) ( ) 1 y T R = x R, R T 1 x T = y R T, A = ( n a T a A ), B = ( ) n b T b B
79 Essentials of Step 2. the quadratic constraints To derive the quadratic constraints, write key ingredients in block form: ( ) ( ) 1 y T R = x R, R T 1 x T = y R T, A = ( n a T a A ), B = ( ) n b T b B This allow several quadratic constraints to be found, eg. det(a xx T ) = a perfect square = det(b yy T )
80 Computational Considerations for Step 2. We need to decide in which order to implement the various quadratic and linear constraints.
81 Computational Considerations for Step 2. We need to decide in which order to implement the various quadratic and linear constraints. How to decide?
82 THE END
Decomposing Gram matrices
Richard P. Brent MSI & CECS ANU Describing joint work with Will Orrick, Judy-anne Osborn and Paul Zimmermann 14 May 2010 Abbreviations maxdet Hadamard maximal determinant problem or a matrix with maximal
More informationExploring Large Graphs
Richard P. Brent MSI & CECS ANU Joint work with Judy-anne Osborn MSI, ANU 1 October 2010 The Hadamard maximal determinant problem The Hadamard maxdet problem asks: what is the maximum determinant H(n)
More informationHadamard and conference matrices
Hadamard and conference matrices Peter J. Cameron December 2011 with input from Dennis Lin, Will Orrick and Gordon Royle Hadamard s theorem Let H be an n n matrix, all of whose entries are at most 1 in
More informationLower bounds on maximal determinants
Lower bounds on maximal determinants Richard P. Brent ANU 26 September 2012 joint work with Judy-anne Osborn University of Newcastle Presented at the 56th annual meeting of the Australian Mathematical
More informationHadamard and conference matrices
Hadamard and conference matrices Peter J. Cameron University of St Andrews & Queen Mary University of London Mathematics Study Group with input from Rosemary Bailey, Katarzyna Filipiak, Joachim Kunert,
More informationHadamard and conference matrices
Hadamard and conference matrices Peter J. Cameron University of St Andrews & Queen Mary University of London Mathematics Study Group with input from Rosemary Bailey, Katarzyna Filipiak, Joachim Kunert,
More informationFinding D-optimal designs by randomised decomposition and switching
AUSTRALASIAN JOURNAL OF COMBINATORICS Volume 55 (2013), Pages 15 30 Finding D-optimal designs by randomised decomposition and switching Richard P. Brent Mathematical Sciences Institute Australian National
More informationThe maximal determinant and subdeterminants of ±1 matrices
Linear Algebra and its Applications 373 (2003) 297 310 www.elsevier.com/locate/laa The maximal determinant and subdeterminants of ±1 matrices Jennifer Seberry a,, Tianbing Xia a, Christos Koukouvinos b,
More informationThe Unimodular Determinant Spectrum Problem
The Unimodular Determinant Spectrum Problem April 18, 2015 What are the possible determinants of ±1 matrices? Order 2 {0, ±2} Order 3 {0, ±4} Order 4 {0, ±8, ±16} Order 5 {0, ±16, ±32, ±48} The Determinant
More informationNeedles, haystacks and optimal designs
Needles, haystacks and optimal designs Richard P. Brent Australian National University and University of Newcastle 15 October 2014 2014 Moyal lecture Introduction In the abstract, I said that I would consider
More informationNew constructions for Hadamard matrices
New constructions for Hadamard matrices Paul Leopardi Mathematical Sciences Institute, Australian National University. For presentation at University of Newcastle. 26 April 2012 Introduction Acknowledgements
More informationAn Overview of Complex Hadamard Cubes
Rose- Hulman Undergraduate Mathematics Journal An Overview of Complex Hadamard Cubes Ben Lantz a Michael Zowada b Volume 3, No. 2, Fall 202 Sponsored by Rose-Hulman Institute of Technology Department of
More informationGeneral lower bounds on maximal determinants of binary matrices
General lower bounds on maximal determinants of binary matrices Richard P. Brent Australian National University Canberra, ACT 0200, Australia maxdet@rpbrent.com Judy-anne H. Osborn The University of Newcastle
More informationTC08 / 6. Hadamard codes SX
TC8 / 6. Hadamard codes 3.2.7 SX Hadamard matrices Hadamard matrices. Paley s construction of Hadamard matrices Hadamard codes. Decoding Hadamard codes A Hadamard matrix of order is a matrix of type whose
More informationSome characterizations of affinely full-dimensional factorial designs
Some characterizations of affinely full-dimensional factorial designs arxiv:0812.0196v1 [stat.me] 1 Dec 2008 Satoshi Aoki and Akimichi Takemura December, 2008 Abstract A new class of two-level non-regular
More informationDifference sets and Hadamard matrices
Difference sets and Hadamard matrices Padraig Ó Catháin National University of Ireland, Galway 14 March 2012 Outline 1 (Finite) Projective planes 2 Symmetric Designs 3 Difference sets 4 Doubly transitive
More informationMath 18.6, Spring 213 Problem Set #6 April 5, 213 Problem 1 ( 5.2, 4). Identify all the nonzero terms in the big formula for the determinants of the following matrices: 1 1 1 2 A = 1 1 1 1 1 1, B = 3 4
More informationON THE CONSTRUCTION OF HADAMARD MATRICES. P.K. Manjhi 1, Arjun Kumar 2. Vinoba Bhave University Hazaribag, INDIA
International Journal of Pure and Applied Mathematics Volume 120 No. 1 2018, 51-58 ISSN: 1311-8080 (printed version); ISSN: 1314-3395 (on-line version) url: http://www.ijpam.eu doi: 10.12732/ijpam.v120i1.4
More information--------------------------------------------------------------------------------------------- Math 6023 Topics: Design and Graph Theory ---------------------------------------------------------------------------------------------
More informationConceptual Questions for Review
Conceptual Questions for Review Chapter 1 1.1 Which vectors are linear combinations of v = (3, 1) and w = (4, 3)? 1.2 Compare the dot product of v = (3, 1) and w = (4, 3) to the product of their lengths.
More informationSupplementary difference sets and optimal designs
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 1991 Supplementary difference sets and optimal designs Christos Koukouvinos
More informationAlgebraic aspects of Hadamard matrices
Algebraic aspects of Hadamard matrices Padraig Ó Catháin University of Queensland 22 February 2013 Overview Difference set Relative difference set Symmetric Design Hadamard matrix Overview 1 Hadamard matrices
More information1. A brief introduction to
1. A brief introduction to design theory These lectures were given to an audience of design theorists; for those outside this class, the introductory chapter describes some of the concepts of design theory
More informationThinking and Doing. Maths in the 21st Century
Thinking and Doing Mathematics in the 21st Century Richard P. Brent MSI, Australian National University and CARMA, University of Newcastle 16 November 2018 Copyright c 2018, R. P. Brent. Maths in the 21st
More informationOn the Least Number of Cell Orbits of a Hadamard Matrix of Order n
Irish Math. Soc. Bulletin 56 (2005), 95 102 95 On the Least Number of Cell Orbits of a Hadamard Matrix of Order n WARWICK DE LAUNEY AND RICHARD STAFFORD Abstract. The automorphism group of any Hadamard
More informationTrades in complex Hadamard matrices
Trades in complex Hadamard matrices Padraig Ó Catháin Ian M. Wanless School of Mathematical Sciences, Monash University, VIC 3800, Australia. February 9, 2015 Abstract A trade in a complex Hadamard matrix
More informationSOME ANALYTICAL ASPECTS OF HADAMARD MATRICES
SOME AALYTICAL ASPECTS OF HADAMARD MATRICES IO ECHITA Contents. The Hadamard conjecture. Historical remarks 2. The determinant of sign matrices 2 3. The l norm of orthogonal matrices 3 4. Almost Hadamard
More informationMA 265 FINAL EXAM Fall 2012
MA 265 FINAL EXAM Fall 22 NAME: INSTRUCTOR S NAME:. There are a total of 25 problems. You should show work on the exam sheet, and pencil in the correct answer on the scantron. 2. No books, notes, or calculators
More informationMath 320, spring 2011 before the first midterm
Math 320, spring 2011 before the first midterm Typical Exam Problems 1 Consider the linear system of equations 2x 1 + 3x 2 2x 3 + x 4 = y 1 x 1 + 3x 2 2x 3 + 2x 4 = y 2 x 1 + 2x 3 x 4 = y 3 where x 1,,
More informationDifference sets and Hadamard matrices
Difference sets and Hadamard matrices Padraig Ó Catháin University of Queensland 5 November 2012 Outline 1 Hadamard matrices 2 Symmetric designs 3 Hadamard matrices and difference sets 4 Two-transitivity
More informationHadamard matrices, difference sets and doubly transitive permutation groups
Hadamard matrices, difference sets and doubly transitive permutation groups Padraig Ó Catháin University of Queensland 13 November 2012 Outline 1 Hadamard matrices 2 Symmetric designs 3 Hadamard matrices
More informationSelf-complementary strongly regular graphs revisited
Self-complementary strongly regular graphs revisited Ferenc Szöllősi (joint work in part with Patric Östergård) szoferi@gmail.com Department of Communications and Networking, Aalto University Talk at CoCoA15
More informationLinear and Bilinear Algebra (2WF04) Jan Draisma
Linear and Bilinear Algebra (2WF04) Jan Draisma CHAPTER 10 Symmetric bilinear forms We already know what a bilinear form β on a K-vector space V is: it is a function β : V V K that satisfies β(u + v,
More informationSome New D-Optimal Designs
Some New D-Optimal Designs Dragomir Z. f) okovic* Department of Pure Mathematics University of Waterloo Waterloo, Ontario N2L 301 Canada E-mail: dragomir@herod.uwaterloo.ca Abstract We construct several
More informationTwo infinite families of symmetric Hadamard matrices. Jennifer Seberry
AUSTRALASIAN JOURNAL OF COMBINATORICS Volume 69(3) (2017), Pages 349 357 Two infinite families of symmetric Hadamard matrices Jennifer Seberry Centre for Computer Security Research School of Computing
More informationarxiv: v4 [math.co] 30 Mar 2009
CYCLIC (v;r,s;λ) DIFFERENCE FAMILIES WITH TWO BASE BLOCKS AND v 50 arxiv:0707.2173v4 [math.co] 30 Mar 2009 DRAGOMIR Ž. D OKOVIĆ Abstract. We construct many new cyclic (v;r,s;λ) difference families with
More informationLecture 1: Latin Squares!
Latin Squares Instructor: Paddy Lecture : Latin Squares! Week of Mathcamp 00 Introduction Definition. A latin square of order n is a n n array, filled with symbols {,... n}, such that no symbol is repeated
More informationLecture 6: Lies, Inner Product Spaces, and Symmetric Matrices
Math 108B Professor: Padraic Bartlett Lecture 6: Lies, Inner Product Spaces, and Symmetric Matrices Week 6 UCSB 2014 1 Lies Fun fact: I have deceived 1 you somewhat with these last few lectures! Let me
More information1 Determinants. 1.1 Determinant
1 Determinants [SB], Chapter 9, p.188-196. [SB], Chapter 26, p.719-739. Bellow w ll study the central question: which additional conditions must satisfy a quadratic matrix A to be invertible, that is to
More informationGaussian Elimination and Back Substitution
Jim Lambers MAT 610 Summer Session 2009-10 Lecture 4 Notes These notes correspond to Sections 31 and 32 in the text Gaussian Elimination and Back Substitution The basic idea behind methods for solving
More informationbesides your solutions of these problems. 1 1 We note, however, that there will be many factors in the admission decision
The PRIMES 2015 Math Problem Set Dear PRIMES applicant! This is the PRIMES 2015 Math Problem Set. Please send us your solutions as part of your PRIMES application by December 1, 2015. For complete rules,
More informationMATH 31 - ADDITIONAL PRACTICE PROBLEMS FOR FINAL
MATH 3 - ADDITIONAL PRACTICE PROBLEMS FOR FINAL MAIN TOPICS FOR THE FINAL EXAM:. Vectors. Dot product. Cross product. Geometric applications. 2. Row reduction. Null space, column space, row space, left
More informationPRODUCTS THAT ARE POWERS. A mathematical vignette Ed Barbeau, University of Toronto
PRODUCTS THAT ARE POWERS. A mathematical vignette Ed Barbeau, University of Toronto This investigation was originally designed for school students to gain fluency with the factoring of positive integers
More informationReview problems for MA 54, Fall 2004.
Review problems for MA 54, Fall 2004. Below are the review problems for the final. They are mostly homework problems, or very similar. If you are comfortable doing these problems, you should be fine on
More information18.S34 linear algebra problems (2007)
18.S34 linear algebra problems (2007) Useful ideas for evaluating determinants 1. Row reduction, expanding by minors, or combinations thereof; sometimes these are useful in combination with an induction
More informationSquare 2-designs/1. 1 Definition
Square 2-designs Square 2-designs are variously known as symmetric designs, symmetric BIBDs, and projective designs. The definition does not imply any symmetry of the design, and the term projective designs,
More informationPROBLEMS ON CONGRUENCES AND DIVISIBILITY
PROBLEMS ON CONGRUENCES AND DIVISIBILITY 1. Do there exist 1,000,000 consecutive integers each of which contains a repeated prime factor? 2. A positive integer n is powerful if for every prime p dividing
More information4.2. ORTHOGONALITY 161
4.2. ORTHOGONALITY 161 Definition 4.2.9 An affine space (E, E ) is a Euclidean affine space iff its underlying vector space E is a Euclidean vector space. Given any two points a, b E, we define the distance
More informationSmall Orders of Hadamard Matrices and Base Sequences
International Mathematical Forum, Vol. 6, 2011, no. 62, 3061-3067 Small Orders of Hadamard Matrices and Base Sequences Dragomir Ž. D oković Department of Pure Mathematics and Institute for Quantum Computing
More information#A30 INTEGERS 17 (2017) CONFERENCE MATRICES WITH MAXIMUM EXCESS AND TWO-INTERSECTION SETS
#A3 INTEGERS 7 (27) CONFERENCE MATRICES WITH MAIMUM ECESS AND TWO-INTERSECTION SETS Koji Momihara Faculty of Education, Kumamoto University, Kurokami, Kumamoto, Japan momihara@educ.kumamoto-u.ac.jp Sho
More informationFinal EXAM Preparation Sheet
Final EXAM Preparation Sheet M369 Fall 217 1 Key concepts The following list contains the main concepts and ideas that we have explored this semester. For each concept, make sure that you remember about
More informationGRE Subject test preparation Spring 2016 Topic: Abstract Algebra, Linear Algebra, Number Theory.
GRE Subject test preparation Spring 2016 Topic: Abstract Algebra, Linear Algebra, Number Theory. Linear Algebra Standard matrix manipulation to compute the kernel, intersection of subspaces, column spaces,
More informationBinary codes of t-designs and Hadamard matrices
Binary codes of t-designs and Hadamard matrices Akihiro Munemasa 1 1 Graduate School of Information Sciences Tohoku University November 8, 2013 JSPS-DST Asian Academic Seminar 2013 Discrete Mathematics
More informationChapter 2. Square matrices
Chapter 2. Square matrices Lecture notes for MA1111 P. Karageorgis pete@maths.tcd.ie 1/18 Invertible matrices Definition 2.1 Invertible matrices An n n matrix A is said to be invertible, if there is a
More informationA Short Overview of Orthogonal Arrays
A Short Overview of Orthogonal Arrays John Stufken Department of Statistics University of Georgia Isaac Newton Institute September 5, 2011 John Stufken (University of Georgia) Orthogonal Arrays September
More informationGroup actions on Hadamard matrices
National University of Ireland, Galway Group actions on Hadamard matrices by Padraig Ó Catháin A thesis submitted in partial fulfillment for the degree of Master of Literature in the Faculty of Arts School
More informationMath 304 Fall 2018 Exam 3 Solutions 1. (18 Points, 3 Pts each part) Let A, B, C, D be square matrices of the same size such that
Math 304 Fall 2018 Exam 3 Solutions 1. (18 Points, 3 Pts each part) Let A, B, C, D be square matrices of the same size such that det(a) = 2, det(b) = 2, det(c) = 1, det(d) = 4. 2 (a) Compute det(ad)+det((b
More informationQ 1 Find the square root of 729. 6. Squares and Square Roots Q 2 Fill in the blank using the given pattern. 7 2 = 49 67 2 = 4489 667 2 = 444889 6667 2 = Q 3 Without adding find the sum of 1 + 3 + 5 + 7
More informationHadamard matrices of order 32
Hadamard matrices of order 32 H. Kharaghani a,1 B. Tayfeh-Rezaie b a Department of Mathematics and Computer Science, University of Lethbridge, Lethbridge, Alberta, T1K3M4, Canada b School of Mathematics,
More informationExercises. Template for Proofs by Mathematical Induction
5. Mathematical Induction 329 Template for Proofs by Mathematical Induction. Express the statement that is to be proved in the form for all n b, P (n) forafixed integer b. 2. Write out the words Basis
More informationGroups and Symmetries
Groups and Symmetries Definition: Symmetry A symmetry of a shape is a rigid motion that takes vertices to vertices, edges to edges. Note: A rigid motion preserves angles and distances. Definition: Group
More information1 Last time: least-squares problems
MATH Linear algebra (Fall 07) Lecture Last time: least-squares problems Definition. If A is an m n matrix and b R m, then a least-squares solution to the linear system Ax = b is a vector x R n such that
More informationComplex Hadamard matrices and some of their applications
Complex Hadamard matrices and some of their applications Karol Życzkowski in collaboration with W. Bruzda and W. S lomczyński (Cracow) W. Tadej (Warsaw) and I. Bengtsson (Stockholm) Institute of Physics,
More informationPROBLEMS ON LINEAR ALGEBRA
1 Basic Linear Algebra PROBLEMS ON LINEAR ALGEBRA 1. Let M n be the (2n + 1) (2n + 1) for which 0, i = j (M n ) ij = 1, i j 1,..., n (mod 2n + 1) 1, i j n + 1,..., 2n (mod 2n + 1). Find the rank of M n.
More informationMath Matrix Algebra
Math 44 - Matrix Algebra Review notes - (Alberto Bressan, Spring 7) sec: Orthogonal diagonalization of symmetric matrices When we seek to diagonalize a general n n matrix A, two difficulties may arise:
More informationMultipartite entangled states, orthogonal arrays & Hadamard matrices. Karol Życzkowski in collaboration with Dardo Goyeneche (Concepcion - Chile)
Multipartite entangled states, orthogonal arrays & Hadamard matrices Karol Życzkowski in collaboration with Dardo Goyeneche (Concepcion - Chile) Institute of Physics, Jagiellonian University, Cracow, Poland
More informationPGSS Discrete Math Solutions to Problem Set #4. Note: signifies the end of a problem, and signifies the end of a proof.
PGSS Discrete Math Solutions to Problem Set #4 Note: signifies the end of a problem, and signifies the end of a proof. 1. Prove that for any k N, there are k consecutive composite numbers. (Hint: (k +
More informationON DIRICHLET S CONJECTURE ON RELATIVE CLASS NUMBER ONE
ON DIRICHLET S CONJECTURE ON RELATIVE CLASS NUMBER ONE AMANDA FURNESS Abstract. We examine relative class numbers, associated to class numbers of quadratic fields Q( m) for m > 0 and square-free. The relative
More informationSolutions to Final Practice Problems Written by Victoria Kala Last updated 12/5/2015
Solutions to Final Practice Problems Written by Victoria Kala vtkala@math.ucsb.edu Last updated /5/05 Answers This page contains answers only. See the following pages for detailed solutions. (. (a x. See
More informationMatrix Factorization and Analysis
Chapter 7 Matrix Factorization and Analysis Matrix factorizations are an important part of the practice and analysis of signal processing. They are at the heart of many signal-processing algorithms. Their
More informationDependencies in discrete tomography. Rob Tijdeman, Milan, Italy, 20 April
Dependencies in discrete tomography Rob Tijdeman, Milan, Italy, 20 April 2015 2013 1. Switching components Let A be a finite set in Z 2. A primitive direction is a relatively prime pair (a, b). Let D be
More informationSudoku and Matrices. Merciadri Luca. June 28, 2011
Sudoku and Matrices Merciadri Luca June 28, 2 Outline Introduction 2 onventions 3 Determinant 4 Erroneous Sudoku 5 Eigenvalues Example 6 Transpose Determinant Trace 7 Antisymmetricity 8 Non-Normality 9
More information[Disclaimer: This is not a complete list of everything you need to know, just some of the topics that gave people difficulty.]
Math 43 Review Notes [Disclaimer: This is not a complete list of everything you need to know, just some of the topics that gave people difficulty Dot Product If v (v, v, v 3 and w (w, w, w 3, then the
More informationNumbers and their divisors
Chapter 1 Numbers and their divisors 1.1 Some number theoretic functions Theorem 1.1 (Fundamental Theorem of Arithmetic). Every positive integer > 1 is uniquely the product of distinct prime powers: n
More informationSome matrices of Williamson type
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 1973 Some matrices of Williamson type Jennifer Seberry University of Wollongong,
More informationRanks of Hadamard Matrices and Equivalence of Sylvester Hadamard and Pseudo-Noise Matrices
Operator Theory: Advances and Applications, Vol 1, 1 13 c 27 Birkhäuser Verlag Basel/Switzerland Ranks of Hadamard Matrices and Equivalence of Sylvester Hadamard and Pseudo-Noise Matrices Tom Bella, Vadim
More informationRandom Lifts of Graphs
27th Brazilian Math Colloquium, July 09 Plan of this talk A brief introduction to the probabilistic method. A quick review of expander graphs and their spectrum. Lifts, random lifts and their properties.
More informationHW2 Solutions Problem 1: 2.22 Find the sign and inverse of the permutation shown in the book (and below).
Teddy Einstein Math 430 HW Solutions Problem 1:. Find the sign and inverse of the permutation shown in the book (and below). Proof. Its disjoint cycle decomposition is: (19)(8)(37)(46) which immediately
More informationFinite geometry codes, generalized Hadamard matrices, and Hamada and Assmus conjectures p. 1/2
Finite geometry codes, generalized Hadamard matrices, and Hamada and Assmus conjectures Vladimir D. Tonchev a Department of Mathematical Sciences Michigan Technological University Houghton, Michigan 49931,
More informationElementary Matrices. which is obtained by multiplying the first row of I 3 by -1, and
Elementary Matrices In this special handout of material not contained in the text, we introduce the concept of elementary matrix. Elementary matrices are useful in several ways that will be shown in this
More informationWarm-up. True or false? Baby proof. 2. The system of normal equations for A x = y has solutions iff A x = y has solutions
Warm-up True or false? 1. proj u proj v u = u 2. The system of normal equations for A x = y has solutions iff A x = y has solutions 3. The normal equations are always consistent Baby proof 1. Let A be
More informationHadamard ideals and Hadamard matrices with two circulant cores
Hadamard ideals and Hadamard matrices with two circulant cores Ilias S. Kotsireas a,1,, Christos Koukouvinos b and Jennifer Seberry c a Wilfrid Laurier University, Department of Physics and Computer Science,
More informationChapter SSM: Linear Algebra Section Fails to be invertible; since det = 6 6 = Invertible; since det = = 2.
SSM: Linear Algebra Section 61 61 Chapter 6 1 2 1 Fails to be invertible; since det = 6 6 = 0 3 6 3 5 3 Invertible; since det = 33 35 = 2 7 11 5 Invertible; since det 2 5 7 0 11 7 = 2 11 5 + 0 + 0 0 0
More informationA REVIEW AND NEW SYMMETRIC CONFERENCE MATRICES
UDC 004438 A REVIEW AND NEW SYMMETRIC CONFERENCE MATRICES N A Balonin a, Dr Sc, Tech, Professor, korbendfs@mailru Jennifer Seberry b, PhD, Foundation Professor, jennifer_seberry@uoweduau a Saint-Petersburg
More informationButson-Hadamard matrices in association schemes of class 6 on Galois rings of characteristic 4
Butson-Hadamard matrices in association schemes of class 6 on Galois rings of characteristic 4 Akihiro Munemasa Tohoku University (joint work with Takuya Ikuta) November 17, 2017 Combinatorics Seminar
More informationSome remarks on Hadamard matrices
University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 2010 Some remarks on Hadamard matrices Jennifer Seberry University of
More informationarxiv: v2 [math.co] 4 Sep 2014
HADAMARD MATRICES MODULO P AND SMALL MODULAR HADAMARD MATRICES arxiv:109.018v [math.co] Sep 01 VIVIAN KUPERBERG ABSTRACT. We use modular symmetric designs to study the existence of Hadamard matrices modulo
More informationOn families of anticommuting matrices
On families of anticommuting matrices Pavel Hrubeš December 18, 214 Abstract Let e 1,..., e k be complex n n matrices such that e ie j = e je i whenever i j. We conjecture that rk(e 2 1) + rk(e 2 2) +
More informationNo books, notes, any calculator, or electronic devices are allowed on this exam. Show all of your steps in each answer to receive a full credit.
MTH 309-001 Fall 2016 Exam 1 10/05/16 Name (Print): PID: READ CAREFULLY THE FOLLOWING INSTRUCTION Do not open your exam until told to do so. This exam contains 7 pages (including this cover page) and 7
More informationMATH CSE20 Homework 5 Due Monday November 4
MATH CSE20 Homework 5 Due Monday November 4 Assigned reading: NT Section 1 (1) Prove the statement if true, otherwise find a counterexample. (a) For all natural numbers x and y, x + y is odd if one of
More informationExam in Linear Algebra
Det danske eksamenssæt findes ved at vende sættet om Exam in Linear Algebra First Year at The Faculties of Engineering and Science and of Health February 7, 206, 9.00-3.00 This test has 9 pages and 5 problems.
More information1 0 1, then use that decomposition to solve the least squares problem. 1 Ax = 2. q 1 = a 1 a 1 = 1. to find the intermediate result:
Exercise Find the QR decomposition of A =, then use that decomposition to solve the least squares problem Ax = 2 3 4 Solution Name the columns of A by A = [a a 2 a 3 ] and denote the columns of the results
More informationSome results on the existence of t-all-or-nothing transforms over arbitrary alphabets
Some results on the existence of t-all-or-nothing transforms over arbitrary alphabets Navid Nasr Esfahani, Ian Goldberg and Douglas R. Stinson David R. Cheriton School of Computer Science University of
More informationMyung-Hwan Kim and Byeong-Kweon Oh. Department of Mathematics, Seoul National University, Seoul , Korea
REPRESENTATIONS OF POSITIVE DEFINITE SENARY INTEGRAL QUADRATIC FORMS BY A SUM OF SQUARES Myung-Hwan Kim and Byeong-Kweon Oh Department of Mathematics Seoul National University Seoul 151-742 Korea Abstract.
More informationA Series of Regular Hadamard Matrices
A Series of Regular Hadamard Matrices Dean Crnković Abstract Let p and p 1 be prime powers and p 3 (mod 4). Then there exists a symmetric design with parameters (4p,p p,p p). Thus there exists a regular
More informationAnnouncements. CS243: Discrete Structures. Sequences, Summations, and Cardinality of Infinite Sets. More on Midterm. Midterm.
Announcements CS43: Discrete Structures Sequences, Summations, and Cardinality of Infinite Sets Işıl Dillig Homework is graded, scores on Blackboard Graded HW and sample solutions given at end of this
More informationPositive definite preserving linear transformations on symmetric matrix spaces
Positive definite preserving linear transformations on symmetric matrix spaces arxiv:1008.1347v1 [math.ra] 7 Aug 2010 Huynh Dinh Tuan-Tran Thi Nha Trang-Doan The Hieu Hue Geometry Group College of Education,
More information2.57 PART E: THE FUNDAMENTAL THEOREM OF ALGEBRA (FTA) The Fundamental Theorem of Algebra (FTA)
2.57 PART E: THE FUNDAMENTAL THEOREM OF ALGEBRA (FTA) The Fundamental Theorem of Algebra (FTA) If f ( x) is a nonconstant n th -degree polynomial in standard form with real coefficients, then it must have
More informationInstitute for Computational Mathematics Hong Kong Baptist University
Institute for Computational Mathematics Hong Kong Baptist University ICM Research Report 08-0 How to find a good submatrix S. A. Goreinov, I. V. Oseledets, D. V. Savostyanov, E. E. Tyrtyshnikov, N. L.
More informationA Brief Outline of Math 355
A Brief Outline of Math 355 Lecture 1 The geometry of linear equations; elimination with matrices A system of m linear equations with n unknowns can be thought of geometrically as m hyperplanes intersecting
More information