An introduction to optimization : with applications to machine learning.: with applications to machine learning. (2023)

• Record Type: Book
• Authors: Chong, Edwin Kah Pin; Lu, Wusheng; Żak, Stanislaw H
• Other Names: Chong, Edwin Kah Pin
• Contents: Preface xv About the Companion Website xviii Part I Mathematical Review 1 1 Methods of Proof and Some Notation 3 1.1 Methods of Proof 3 1.2 Notation 5 Exercises 5 2 Vector Spaces and Matrices 7 2.1 Vector and Matrix 7 2.2 Rank of a Matrix 11 2.3 Linear Equations 16 2.4 Inner Products and Norms 18 Exercises 20 3 Transformations 23 3.1 Linear Transformations 23 3.2 Eigenvalues and Eigenvectors 24 3.3 Orthogonal Projections 26 3.4 Quadratic Forms 27 3.5 Matrix Norms 32 Exercises 35 4 Concepts from Geometry 39 4.1 Line Segments 39 4.2 Hyperplanes and Linear Varieties 39 4.3 Convex Sets 41 4.4 Neighborhoods 43 4.5 Polytopes and Polyhedra 44 Exercises 45 5 Elements of Calculus 47 5.1 Sequences and Limits 47 5.2 Differentiability 52 5.3 The Derivative Matrix 54 5.4 Differentiation Rules 57 5.5 Level Sets and Gradients 58 5.6 Taylor Series 61 Exercises 65 Part II Unconstrained Optimization 67 6 Basics of Set-Constrained and Unconstrained Optimization 69 6.1 Introduction 69 6.2 Conditions for Local Minimizers 70 Exercises 78 7 One-Dimensional Search Methods 87 7.1 Introduction 87 7.2 Golden Section Search 87 7.3 Fibonacci Method 91 7.4 Bisection Method 97 7.5 Newton’s Method 98 7.6 Secant Method 101 7.7 Bracketing 103 7.8 Line Search in Multidimensional Optimization 103 Exercises 105 8 Gradient Methods 109 8.1 Introduction 109 8.2 Steepest Descent Method 110 8.3 Analysis of Gradient Methods 117 Exercises 126 9 Newton’s Method 133 9.1 Introduction 133 9.2 Analysis of Newton’s Method 135 9.3 Levenberg–Marquardt Modification 138 9.4 Newton’s Method for Nonlinear Least Squares 139 Exercises 142 10 Conjugate Direction Methods 145 10.1 Introduction 145 10.2 Conjugate Direction Algorithm 146 10.2.1 Basic Conjugate Direction Algorithm 146 10.3 Conjugate Gradient Algorithm 151 10.4 Conjugate Gradient Algorithm for Nonquadratic Problems 154 Exercises 156 11 Quasi-Newton Methods 159 11.1 Introduction 159 11.2 Approximating the Inverse Hessian 160 11.3 Rank One Correction Formula 162 11.4 DFP Algorithm 166 11.5 BFGS Algorithm 170 Exercises 173 12 Solving Linear Equations 179 12.1 Least-Squares Analysis 179 12.2 Recursive Least-Squares Algorithm 187 12.3 Solution to a Linear Equation with Minimum Norm 190 12.4 Kaczmarz’s Algorithm 191 12.5 Solving Linear Equations in General 194 Exercises 201 13 Unconstrained Optimization and Neural Networks 209 13.1 Introduction 209 13.2 Single-Neuron Training 211 13.3 Backpropagation Algorithm 213 Exercises 222 14 Global Search Algorithms 225 14.1 Introduction 225 14.2 Nelder–Mead Simplex Algorithm 225 14.3 Simulated Annealing 229 14.3.1 Randomized Search 229 14.3.2 Simulated Annealing Algorithm 229 14.4 Particle Swarm Optimization 231 14.4.1 Basic PSO Algorithm 232 14.4.2 Variations 233 14.5 Genetic Algorithms 233 14.5.1 Basic Description 233 14.5.1.1 Chromosomes and Representation Schemes 234 14.5.1.2 Selection and Evolution 234 14.5.2 Analysis of Genetic Algorithms 238 14.5.3 Real-Number Genetic Algorithms 243 Exercises 244 Part III Linear Programming 247 15 Introduction to Linear Programming 249 15.1 Brief History of Linear Programming 249 15.2 Simple Examples of Linear Programs 250 15.3 Two-Dimensional Linear Programs 256 15.4 Convex Polyhedra and Linear Programming 258 15.5 Standard Form Linear Programs 260 15.6 Basic Solutions 264 15.7 Properties of Basic Solutions 267 15.8 Geometric View of Linear Programs 269 Exercises 273 16 Simplex Method 277 16.1 Solving Linear Equations Using Row Operations 277 16.2 The Canonical Augmented Matrix 283 16.3 Updating the Augmented Matrix 284 16.4 The Simplex Algorithm 285 16.5 Matrix Form of the Simplex Method 291 16.6 Two-Phase Simplex Method 294 16.7 Revised Simplex Method 297 Exercises 301 17 Duality 309 17.1 Dual Linear Programs 309 17.2 Properties of Dual Problems 316 17.3 Matrix Games 321 Exercises 324 18 Nonsimplex Methods 331 18.1 Introduction 331 18.2 Khachiyan’s Method 332 18.3 Affine Scaling Method 334 18.3.1 Basic Algorithm 334 18.3.2 Two-Phase Method 337 18.4 Karmarkar’s Method 339 18.4.1 Basic Ideas 339 18.4.2 Karmarkar’s Canonical Form 339 18.4.3 Karmarkar’s Restricted Problem 341 18.4.4 From General Form to Karmarkar’s Canonical Form 342 18.4.5 The Algorithm 345 Exercises 349 19 Integer Linear Programming 351 19.1 Introduction 351 19.2 Unimodular Matrices 351 19.3 The Gomory Cutting-Plane Method 358 Exercises 366 Part IV Nonlinear Constrained Optimization 369 20 Problems with Equality Constraints 371 20.1 Introduction 371 20.2 Problem Formulation 373 20.3 Tangent and Normal Spaces 374 20.4 Lagrange Condition 379 20.5 Second-Order Conditions 387 20.6 Minimizing Quadratics Subject to Linear Constraints 390 Exercises 394 21 Problems with Inequality Constraints 399 21.1 Karush–Kuhn–Tucker Condition 399 21.2 Second-Order Conditions 406 Exercises 410 22 Convex Optimization Problems 417 22.1 Introduction 417 22.2 Convex Functions 419 22.3 Convex Optimization Problems 426 22.4 Semidefinite Programming 431 22.4.1 Linear Matrix Inequalities and Their Properties 431 22.4.2 LMI Solvers 435 22.4.2.1 Finding a Feasible Solution Under LMI Constraints 436 22.4.2.2 Minimizing a Linear Objective Under LMI Constraints 438 22.4.2.3 Minimizing a Generalized Eigenvalue Under LMI Constraints 440 Exercises 442 23 Lagrangian Duality 449 23.1 Overview 449 23.2 Notation 449 23.3 Primal–Dual Pair 450 23.4 General Duality Properties 451 23.4.1 Convexity of Dual Problem 451 23.4.2 Primal Objective in Terms of Lagrangian 451 23.4.3 Minimax Inequality Chain 452 23.4.4 Optimality of Saddle Point 452 23.4.5 Weak Duality 453 23.4.6 Duality Gap 453 23.5 Strong Duality 454 23.5.1 Strong Duality ⇔ Minimax Equals Maximin 454 23.5.2 Strong Duality ⇒ Primal Unconstrained Minimization 455 23.5.3 Strong Duality ⇒ Optimality 455 23.5.4 Strong Duality ⇒ KKT (Including Complementary Slackness) 455 23.5.5 Strong Duality ⇒ Saddle Point 456 23.6 Con
• Edition: Fifth edition
• Publisher Details: Hoboken : John Wiley & Sons, Inc
• Publication Date: 2023
• Extent: 1 online resource (672 pages), illustrations (black and white)
• Subjects: 519.6; Mathematical optimization
• Languages: English
• ISBNs: 9781119877653
• Notes: Note: Includes bibliographical references and index.; Note: Description based on CIP data; resource not viewed.
• Access Rights: Legal Deposit; Only available on premises controlled by the deposit library and to one user at any one time; The Legal Deposit Libraries (Non-Print Works) Regulations (UK).
• Access Usage: Restricted: Printing from this resource is governed by The Legal Deposit Libraries (Non-Print Works) Regulations (UK) and UK copyright law currently in force.