DOCSLIB.ORG

Discover Linear Algebra Incomplete Preliminary Draft

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Discover Linear Algebra Incomplete Preliminary Draft Discover Linear Algebra Incomplete Preliminary Draft Date: November 28, 2017 L´aszl´oBabai in collaboration with Noah Halford All rights reserved. Approved for instructional use only. Commercial distribution prohibited. c 2016 L´aszl´oBabai. Last updated: November 10, 2016 Preface TO BE WRITTEN. Babai: Discover Linear Algebra. ii This chapter last updated August 21, 2016 c 2016 L´aszl´oBabai. Contents Notation ix I Matrix Theory 1 Introduction to Part I 2 1 (F, R) Column Vectors 3 1.1 (F) Column vector basics . 3 1.1.1 The domain of scalars . 3 1.2 (F) Subspaces and span . 6 1.3 (F) Linear independence and the First Miracle of Linear Algebra . 8 1.4 (F) Dot product . 12 1.5 (R) Dot product over R ................................. 14 1.6 (F) Additional exercises . 14 2 (F) Matrices 15 2.1 Matrix basics . 15 2.2 Matrix multiplication . 18 2.3 Arithmetic of diagonal and triangular matrices . 22 2.4 Permutation Matrices . 24 2.5 Additional exercises . 26 3 (F) Matrix Rank 28 3.1 Column and row rank . 28 iii iv CONTENTS 3.2 Elementary operations and Gaussian elimination . 29 3.3 Invariance of column and row rank, the Second Miracle of Linear Algebra . 31 3.4 Matrix rank and invertibility . 33 3.5 Codimension (optional) . 34 3.6 Additional exercises . 35 4 (F) Theory of Systems of Linear Equations I: Qualitative Theory 38 4.1 Homogeneous systems of linear equations . 38 4.2 General systems of linear equations . 40 5 (F, R) Affine and Convex Combinations (optional) 42 5.1 (F) Affine combinations . 42 5.2 (F) Hyperplanes . 44 5.3 (R) Convex combinations . 45 5.4 (R) Helly's Theorem . 46 5.5 (F, R) Additional exercises . 47 6 (F) The Determinant 48 6.1 Permutations . 49 6.2 Defining the determinant . 51 6.3 Cofactor expansion . 54 6.4 Matrix inverses and the determinant . 55 6.5 Additional exercises . 56 7 (F) Theory of Systems of Linear Equations II: Cramer's Rule 57 7.1 Cramer's Rule . 57 8 (F) Eigenvectors and Eigenvalues 59 8.1 Eigenvector and eigenvalue basics . 59 8.2 Similar matrices and diagonalizability . 61 8.3 Polynomial basics . 61 8.4 The characteristic polynomial . 63 8.5 The Cayley-Hamilton Theorem . 64 8.6 Additional exercises . 66 CONTENTS v 9 (R) Orthogonal Matrices 67 9.1 Orthogonal matrices . 67 9.2 Orthogonal similarity . 67 9.3 Additional exercises . 68 10 (R) The Spectral Theorem 69 10.1 Statement of the Spectral Theorem . 69 10.2 Applications of the Spectral Theorem . 69 11 (F, R) Bilinear and Quadratic Forms 71 11.1 (F) Linear and bilinear forms . 71 11.2 (F) Multivariate polynomials . 72 11.3 (R) Quadratic forms . 74 11.4 (F) Geometric algebra (optional) . 76 12 (C) Complex Matrices 78 12.1 Complex numbers . 78 12.2 Hermitian dot product in Cn .............................. 78 12.3 Hermitian and unitary matrices . 80 12.4 Normal matrices and unitary similarity . 81 12.5 Additional exercises . 82 13 (C, R) Matrix Norms 83 13.1 (R) Operator norm . 83 13.2 (R) Frobenius norm . 84 13.3 (C) Complex Matrices . 84 II Linear Algebra of Vector Spaces 85 Introduction to Part II 86 14 (F) Preliminaries 87 14.1 Modular arithmetic . 87 14.2 Abelian groups . 90 14.3 Fields . 93 vi CONTENTS 14.4 Polynomials . 94 15 (F) Vector Spaces: Basic Concepts 102 15.1 Vector spaces . 102 15.2 Subspaces and span . 104 15.3 Linear independence and bases . 105 15.4 The First Miracle of Linear Algebra . 109 15.5 Direct sums . 111 16 (F) Linear Maps 112 16.1 Linear map basics . 112 16.2 Isomorphisms . 113 16.3 The Rank-Nullity Theorem . 114 16.4 Linear transformations . 115 16.4.1 Eigenvectors, eigenvalues, eigenspaces . 116 16.4.2 Invariant subspaces . 117 16.5 Coordinatization . 120 16.6 Change of basis . 122 17 (F) Block Matrices (optional) 124 17.1 Block matrix basics . 124 17.2 Arithmetic of block-diagonal and block-triangular matrices . 125 18 (F) Minimal Polynomials of Matrices and Linear Transformations (optional) 127 18.1 The minimal polynomial . 127 18.2 Minimal polynomials of linear transformations . 128 19 (R) Euclidean Spaces 131 19.1 Inner products . 131 19.2 Gram-Schmidt orthogonalization . 134 19.3 Isometries and orthogonal transformations . 136 19.4 First proof of the Spectral Theorem . 138 20 (C) Hermitian Spaces 141 20.1 Hermitian spaces . 141 20.2 Hermitian transformations . 144 CONTENTS vii 20.3 Unitary transformations . 145 20.4 Adjoint transformations in Hermitian spaces . 146 20.5 Normal transformations . 146 20.6 The Complex Spectral Theorem for normal transformations . 147 21 (R, C) The Singular Value Decomposition 148 21.1 The Singular Value Decomposition . 148 21.2 Low-rank approximation . 149 22 (R) Finite Markov Chains 151 22.1 Stochastic matrices . 151 22.2 Finite Markov Chains . 151 22.3 Digraphs . 153 22.4 Digraphs and Markov Chains . 153 22.5 Finite Markov Chains and undirected graphs . 153 22.6 Additional exercises . 153 23 More Chapters 155 24 Hints 156 24.1 Column Vectors . ..
Load more

Recommended publications
  • Linear Independence, Span, and Basis of a Set of Vectors What Is Linear Independence?
    LECTURENOTES · purdue university MA 26500 Kyle Kloster Linear Algebra October 22, 2014 Linear Independence, Span, and Basis of a Set of Vectors What is linear independence? A set of vectors S = fv1; ··· ; vkg is linearly independent if none of the vectors vi can be written as a linear combination of the other vectors, i.e. vj = α1v1 + ··· + αkvk. Suppose the vector vj can be written as a linear combination of the other vectors, i.e. there exist scalars αi such that vj = α1v1 + ··· + αkvk holds. (This is equivalent to saying that the vectors v1; ··· ; vk are linearly dependent). We can subtract vj to move it over to the other side to get an expression 0 = α1v1 + ··· αkvk (where the term vj now appears on the right hand side. In other words, the condition that \the set of vectors S = fv1; ··· ; vkg is linearly dependent" is equivalent to the condition that there exists αi not all of which are zero such that 2 3 α1 6α27 0 = v v ··· v 6 7 : 1 2 k 6 . 7 4 . 5 αk More concisely, form the matrix V whose columns are the vectors vi. Then the set S of vectors vi is a linearly dependent set if there is a nonzero solution x such that V x = 0. This means that the condition that \the set of vectors S = fv1; ··· ; vkg is linearly independent" is equivalent to the condition that \the only solution x to the equation V x = 0 is the zero vector, i.e. x = 0. How do you determine if a set is lin.
    [Show full text]
  • Linear Independence, the Wronskian, and Variation of Parameters
    LINEAR INDEPENDENCE, THE WRONSKIAN, AND VARIATION OF PARAMETERS JAMES KEESLING In this post we determine when a set of solutions of a linear differential equation are linearly independent. We first discuss the linear space of solutions for a homogeneous differential equation. 1. Homogeneous Linear Differential Equations We start with homogeneous linear nth-order ordinary differential equations with general coefficients. The form for the nth-order type of equation is the following. dnx dn−1x (1) a (t) + a (t) + ··· + a (t)x = 0 n dtn n−1 dtn−1 0 It is straightforward to solve such an equation if the functions ai(t) are all constants. However, for general functions as above, it may not be so easy. However, we do have a principle that is useful. Because the equation is linear and homogeneous, if we have a set of solutions fx1(t); : : : ; xn(t)g, then any linear combination of the solutions is also a solution. That is (2) x(t) = C1x1(t) + C2x2(t) + ··· + Cnxn(t) is also a solution for any choice of constants fC1;C2;:::;Cng. Now if the solutions fx1(t); : : : ; xn(t)g are linearly independent, then (2) is the general solution of the differential equation. We will explain why later. What does it mean for the functions, fx1(t); : : : ; xn(t)g, to be linearly independent? The simple straightforward answer is that (3) C1x1(t) + C2x2(t) + ··· + Cnxn(t) = 0 implies that C1 = 0, C2 = 0, ::: , and Cn = 0 where the Ci's are arbitrary constants. This is the definition, but it is not so easy to determine from it just when the condition holds to show that a given set of functions, fx1(t); x2(t); : : : ; xng, is linearly independent.
    [Show full text]
  • Span, Linear Independence and Basis Rank and Nullity
    Remarks for Exam 2 in Linear Algebra Span, linear independence and basis The span of a set of vectors is the set of all linear combinations of the vectors. A set of vectors is linearly independent if the only solution to c1v1 + ::: + ckvk = 0 is ci = 0 for all i. Given a set of vectors, you can determine if they are linearly independent by writing the vectors as the columns of the matrix A, and solving Ax = 0. If there are any non-zero solutions, then the vectors are linearly dependent. If the only solution is x = 0, then they are linearly independent. A basis for a subspace S of Rn is a set of vectors that spans S and is linearly independent. There are many bases, but every basis must have exactly k = dim(S) vectors. A spanning set in S must contain at least k vectors, and a linearly independent set in S can contain at most k vectors. A spanning set in S with exactly k vectors is a basis. A linearly independent set in S with exactly k vectors is a basis. Rank and nullity The span of the rows of matrix A is the row space of A. The span of the columns of A is the column space C(A). The row and column spaces always have the same dimension, called the rank of A. Let r = rank(A). Then r is the maximal number of linearly independent row vectors, and the maximal number of linearly independent column vectors. So if r < n then the columns are linearly dependent; if r < m then the rows are linearly dependent.
    [Show full text]
  • Download Special Issue
    Scientific Programming Scientific Programming Techniques and Algorithms for Data-Intensive Engineering Environments Lead Guest Editor: Giner Alor-Hernandez Guest Editors: Jezreel Mejia-Miranda and José María Álvarez-Rodríguez Scientific Programming Techniques and Algorithms for Data-Intensive Engineering Environments Scientific Programming Scientific Programming Techniques and Algorithms for Data-Intensive Engineering Environments Lead Guest Editor: Giner Alor-Hernández Guest Editors: Jezreel Mejia-Miranda and José María Álvarez-Rodríguez Copyright © 2018 Hindawi. All rights reserved. This is a special issue published in “Scientific Programming.” All articles are open access articles distributed under the Creative Com- mons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Editorial Board M. E. Acacio Sanchez, Spain Christoph Kessler, Sweden Danilo Pianini, Italy Marco Aldinucci, Italy Harald Köstler, Germany Fabrizio Riguzzi, Italy Davide Ancona, Italy José E. Labra, Spain Michele Risi, Italy Ferruccio Damiani, Italy Thomas Leich, Germany Damian Rouson, USA Sergio Di Martino, Italy Piotr Luszczek, USA Giuseppe Scanniello, Italy Basilio B. Fraguela, Spain Tomàs Margalef, Spain Ahmet Soylu, Norway Carmine Gravino, Italy Cristian Mateos, Argentina Emiliano Tramontana, Italy Gianluigi Greco, Italy Roberto Natella, Italy Autilia Vitiello, Italy Bormin Huang, USA Francisco Ortin, Spain Jan Weglarz, Poland Chin-Yu Huang, Taiwan Can Özturan, Turkey
    [Show full text]
  • Matroidal Subdivisions, Dressians and Tropical Grassmannians
    Matroidal subdivisions, Dressians and tropical Grassmannians vorgelegt von Diplom-Mathematiker Benjamin Frederik Schröter geboren in Frankfurt am Main Von der Fakultät II – Mathematik und Naturwissenschaften der Technischen Universität Berlin zur Erlangung des akademischen Grades Doktor der Naturwissenschaften – Dr. rer. nat. – genehmigte Dissertation Promotionsausschuss: Vorsitzender: Prof. Dr. Wilhelm Stannat Gutachter: Prof. Dr. Michael Joswig Prof. Dr. Hannah Markwig Senior Lecturer Ph.D. Alex Fink Tag der wissenschaftlichen Aussprache: 17. November 2017 Berlin 2018 Zusammenfassung In dieser Arbeit untersuchen wir verschiedene Aspekte von tropischen linearen Räumen und deren Modulräumen, den tropischen Grassmannschen und Dressschen. Tropische lineare Räume sind dual zu Matroidunterteilungen. Motiviert durch das Konzept der Splits, dem einfachsten Fall einer polytopalen Unterteilung, wird eine neue Klasse von Matroiden eingeführt, die mit Techniken der polyedrischen Geometrie untersucht werden kann. Diese Klasse ist sehr groß, da sie alle Paving-Matroide und weitere Matroide enthält. Die strukturellen Eigenschaften von Split-Matroiden können genutzt werden, um neue Ergebnisse in der tropischen Geometrie zu erzielen. Vor allem verwenden wir diese, um Strahlen der tropischen Grassmannschen zu konstruieren und die Dimension der Dressschen zu bestimmen. Dazu wird die Beziehung zwischen der Realisierbarkeit von Matroiden und der von tropischen linearen Räumen weiter entwickelt. Die Strahlen einer Dressschen entsprechen den Facetten des Sekundärpolytops eines Hypersimplexes. Eine besondere Klasse von Facetten bildet die Verallgemeinerung von Splits, die wir Multi-Splits nennen und die Herrmann ursprünglich als k-Splits bezeichnet hat. Wir geben eine explizite kombinatorische Beschreibung aller Multi-Splits eines Hypersimplexes. Diese korrespondieren mit Nested-Matroiden. Über die tropische Stiefelabbildung erhalten wir eine Beschreibung aller Multi-Splits für Produkte von Simplexen.
    [Show full text]
  • 1 Sets and Set Notation. Definition 1 (Naive Definition of a Set)
    LINEAR ALGEBRA MATH 2700.006 SPRING 2013 (COHEN) LECTURE NOTES 1 Sets and Set Notation. Definition 1 (Naive Definition of a Set). A set is any collection of objects, called the elements of that set. We will most often name sets using capital letters, like A, B, X, Y , etc., while the elements of a set will usually be given lower-case letters, like x, y, z, v, etc. Two sets X and Y are called equal if X and Y consist of exactly the same elements. In this case we write X = Y . Example 1 (Examples of Sets). (1) Let X be the collection of all integers greater than or equal to 5 and strictly less than 10. Then X is a set, and we may write: X = f5; 6; 7; 8; 9g The above notation is an example of a set being described explicitly, i.e. just by listing out all of its elements. The set brackets {· · ·} indicate that we are talking about a set and not a number, sequence, or other mathematical object. (2) Let E be the set of all even natural numbers. We may write: E = f0; 2; 4; 6; 8; :::g This is an example of an explicity described set with infinitely many elements. The ellipsis (:::) in the above notation is used somewhat informally, but in this case its meaning, that we should \continue counting forever," is clear from the context. (3) Let Y be the collection of all real numbers greater than or equal to 5 and strictly less than 10. Recalling notation from previous math courses, we may write: Y = [5; 10) This is an example of using interval notation to describe a set.
    [Show full text]
  • Memory Capacity of Neural Turing Machines with Matrix Representation
    Memory Capacity of Neural Turing Machines with Matrix Representation Animesh Renanse * [email protected] Department of Electronics & Electrical Engineering Indian Institute of Technology Guwahati Assam, India Rohitash Chandra * [email protected] School of Mathematics and Statistics University of New South Wales Sydney, NSW 2052, Australia Alok Sharma [email protected] Laboratory for Medical Science Mathematics RIKEN Center for Integrative Medical Sciences Yokohama, Japan *Equal contributions Editor: Abstract It is well known that recurrent neural networks (RNNs) faced limitations in learning long- term dependencies that have been addressed by memory structures in long short-term memory (LSTM) networks. Matrix neural networks feature matrix representation which inherently preserves the spatial structure of data and has the potential to provide better memory structures when compared to canonical neural networks that use vector repre- sentation. Neural Turing machines (NTMs) are novel RNNs that implement notion of programmable computers with neural network controllers to feature algorithms that have copying, sorting, and associative recall tasks. In this paper, we study augmentation of mem- arXiv:2104.07454v1 [cs.LG] 11 Apr 2021 ory capacity with matrix representation of RNNs and NTMs (MatNTMs). We investigate if matrix representation has a better memory capacity than the vector representations in conventional neural networks. We use a probabilistic model of the memory capacity us- ing Fisher information and investigate how the memory capacity for matrix representation networks are limited under various constraints, and in general, without any constraints. In the case of memory capacity without any constraints, we found that the upper bound on memory capacity to be N 2 for an N ×N state matrix.
    [Show full text]
  • Homogeneous Systems (1.5) Linear Independence and Dependence (1.7)
    Math 20F, 2015SS1 / TA: Jor-el Briones / Sec: A01 / Handout Page 1 of3 Homogeneous systems (1.5) Homogenous systems are linear systems in the form Ax = 0, where 0 is the 0 vector. Given a system Ax = b, suppose x = α + t1α1 + t2α2 + ::: + tkαk is a solution (in parametric form) to the system, for any values of t1; t2; :::; tk. Then α is a solution to the system Ax = b (seen by seeting t1 = ::: = tk = 0), and α1; α2; :::; αk are solutions to the homoegeneous system Ax = 0. Terms you should know The zero solution (trivial solution): The zero solution is the 0 vector (a vector with all entries being 0), which is always a solution to the homogeneous system Particular solution: Given a system Ax = b, suppose x = α + t1α1 + t2α2 + ::: + tkαk is a solution (in parametric form) to the system, α is the particular solution to the system. The other terms constitute the solution to the associated homogeneous system Ax = 0. Properties of a Homegeneous System 1. A homogeneous system is ALWAYS consistent, since the zero solution, aka the trivial solution, is always a solution to that system. 2. A homogeneous system with at least one free variable has infinitely many solutions. 3. A homogeneous system with more unknowns than equations has infinitely many solu- tions 4. If α1; α2; :::; αk are solutions to a homogeneous system, then ANY linear combination of α1; α2; :::; αk is also a solution to the homogeneous system Important theorems to know Theorem. (Chapter 1, Theorem 6) Given a consistent system Ax = b, suppose α is a solution to the system.
    [Show full text]
  • Linear Algebra Review
    Linear Algebra Review Kaiyu Zheng October 2017 Linear algebra is fundamental for many areas in computer science. This document aims at providing a reference (mostly for myself) when I need to remember some concepts or examples. Instead of a collection of facts as the Matrix Cookbook, this document is more gentle like a tutorial. Most of the content come from my notes while taking the undergraduate linear algebra course (Math 308) at the University of Washington. Contents on more advanced topics are collected from reading different sources on the Internet. Contents 3.8 Exponential and 7 Special Matrices 19 Logarithm...... 11 7.1 Block Matrix.... 19 1 Linear System of Equa- 3.9 Conversion Be- 7.2 Orthogonal..... 20 tions2 tween Matrix Nota- 7.3 Diagonal....... 20 tion and Summation 12 7.4 Diagonalizable... 20 2 Vectors3 7.5 Symmetric...... 21 2.1 Linear independence5 4 Vector Spaces 13 7.6 Positive-Definite.. 21 2.2 Linear dependence.5 4.1 Determinant..... 13 7.7 Singular Value De- 2.3 Linear transforma- 4.2 Kernel........ 15 composition..... 22 tion.........5 4.3 Basis......... 15 7.8 Similar........ 22 7.9 Jordan Normal Form 23 4.4 Change of Basis... 16 3 Matrix Algebra6 7.10 Hermitian...... 23 4.5 Dimension, Row & 7.11 Discrete Fourier 3.1 Addition.......6 Column Space, and Transform...... 24 3.2 Scalar Multiplication6 Rank......... 17 3.3 Matrix Multiplication6 8 Matrix Calculus 24 3.4 Transpose......8 5 Eigen 17 8.1 Differentiation... 24 3.4.1 Conjugate 5.1 Multiplicity of 8.2 Jacobian......
    [Show full text]
  • A SHORT PROOF of the EXISTENCE of JORDAN NORMAL FORM Let V Be a Finite-Dimensional Complex Vector Space and Let T : V → V Be A
    A SHORT PROOF OF THE EXISTENCE OF JORDAN NORMAL FORM MARK WILDON Let V be a finite-dimensional complex vector space and let T : V → V be a linear map. A fundamental theorem in linear algebra asserts that there is a basis of V in which T is represented by a matrix in Jordan normal form J1 0 ... 0 0 J2 ... 0 . . .. . 0 0 ...Jk where each Ji is a matrix of the form λ 1 ... 0 0 0 λ . 0 0 . . .. 0 0 . λ 1 0 0 ... 0 λ for some λ ∈ C. We shall assume that the usual reduction to the case where some power of T is the zero map has been made. (See [1, §58] for a characteristically clear account of this step.) After this reduction, it is sufficient to prove the following theorem. Theorem 1. If T : V → V is a linear transformation of a finite-dimensional vector space such that T m = 0 for some m ≥ 1, then there is a basis of V of the form a1−1 ak−1 u1, T u1,...,T u1, . , uk, T uk,...,T uk a where T i ui = 0 for 1 ≤ i ≤ k. At this point all the proofs the author has seen (even Halmos’ in [1, §57]) become unnecessarily long-winded. In this note we present a simple proof which leads to a straightforward algorithm for finding the required basis. Date: December 2007. 1 2 MARK WILDON Proof. We work by induction on dim V . For the inductive step we may assume that dim V ≥ 1.
    [Show full text]
  • Noncommutative Spaces from Matrix Models Lei Lu Allen B Stern, Committee Chair Nobuchika Okada Benjamin Harms Gary Mankey Martyn
    NONCOMMUTATIVE SPACES FROM MATRIX MODELS by LEI LU ALLEN B STERN, COMMITTEE CHAIR NOBUCHIKA OKADA BENJAMIN HARMS GARY MANKEY MARTYN DIXON A DISSERTATION Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Physics and Astronomy in the Graduate School of The University of Alabama TUSCALOOSA, ALABAMA 2016 Copyright Lei Lu 2016 ALL RIGHTS RESERVED Abstract Noncommutative (NC) spaces commonly arise as solutions to matrix model equations of motion. They are natural generalizations of the ordinary commuta- tive spacetime. Such spaces may provide insights into physics close to the Planck scale, where quantum gravity becomes relevant. Although there has been much research in the literature, aspects of these NC spaces need further investigation. In this dissertation, we focus on properties of NC spaces in several different contexts. In particular, we study exact NC spaces which result from solutions to matrix model equations of motion. These spaces are associated with finite-dimensional Lie-algebras. More specifically, they are two-dimensional fuzzy spaces that arise from a three-dimensional Yang-Mills type matrix model, four-dimensional tensor- product fuzzy spaces from a tensorial matrix model, and Snyder algebra from a five-dimensional tensorial matrix model. In the first part of this dissertation, we study two-dimensional NC solutions to matrix equations of motion of extended IKKT-type matrix models in three- space-time dimensions. Perturbations around the NC solutions lead to NC field theories living on a two-dimensional space-time. The commutative limit of the solutions are smooth manifolds which can be associated with closed, open and static two-dimensional cosmologies.
    [Show full text]
  • Linear Algebra (VII)
    Linear Algebra (VII) Yijia Chen 1. Review Basis and Dimension. We fix a vector space V. Lemma 1.1. Let A, B ⊆ V be two finite sets of vectors in V, possibly empty. If A is linearly indepen- dent and can be represented by B. Then jAj 6 jBj. Theorem 1.2. Let S ⊆ V and A, B ⊆ S be both maximally linearly independent in S. Then jAj = jBj. Definition 1.3. Let e1,..., en 2 V. Assume that – e1,..., en are linearly independent, – and every v 2 V can be represented by e1,..., en. Equivalently, fe1,..., eng is maximally linearly independent in V. Then fe1,..., eng is a basis of V. Note that n = 0 is allowed, and in that case, it is easy to see that V = f0g. By Theorem 1.2: 0 0 Lemma 1.4. If fe1,..., eng and fe1,..., emg be both bases of V with pairwise distinct ei’s and with 0 pairwise distinct ei, then n = m. Definition 1.5. Let fe1,..., eng be a basis of V with pairwise distinct ei’s. Then the dimension of V, denoted by dim(V), is n. Equivalently, if rank(V) is defined, then dim(V) := rank(V). 1 Theorem 1.6. Assume dim(V) = n and let u1,..., un 2 V. (1) If u1,..., un are linearly independent, then fu1,..., ung is a basis. (2) If every v 2 V can be represented by u1,..., un, then fu1,..., ung is a basis. Steinitz exchange lemma. Theorem 1.7. Assume that dim(V) = n and v1,..., vm 2 V with 1 6 m 6 n are linearly indepen- dent.
    [Show full text]