Krylov-Subspace Methods - I Lecture 6 Alessandra Nardi Thanks to Prof. Jacob White, Deepak Ramaswamy, Michal Rewienski, and Karen Veroy.

Slides:



Advertisements
Similar presentations
10.4 Complex Vector Spaces.
Advertisements

3D Geometry for Computer Graphics
Scientific Computing QR Factorization Part 2 – Algorithm to Find Eigenvalues.
Optimization.
Solving Linear Systems (Numerical Recipes, Chap 2)
OCE301 Part II: Linear Algebra lecture 4. Eigenvalue Problem Ax = y Ax = x occur frequently in engineering analysis (eigenvalue problem) Ax =  x [ A.
Iterative Methods and QR Factorization Lecture 5 Alessandra Nardi Thanks to Prof. Jacob White, Suvranu De, Deepak Ramaswamy, Michal Rewienski, and Karen.
Steepest Decent and Conjugate Gradients (CG). Solving of the linear equation system.
Modern iterative methods For basic iterative methods, converge linearly Modern iterative methods, converge faster –Krylov subspace method Steepest descent.
1cs542g-term High Dimensional Data  So far we’ve considered scalar data values f i (or interpolated/approximated each component of vector values.
Jonathan Richard Shewchuk Reading Group Presention By David Cline
Symmetric Matrices and Quadratic Forms
Computer Graphics Recitation 5.
QR Factorization –Direct Method to solve linear systems Problems that generate Singular matrices –Modified Gram-Schmidt Algorithm –QR Pivoting Matrix must.
Lecture 12 Least Square Approximation Shang-Hua Teng.
Lecture 13 Operations in Graphics and Geometric Modeling I: Projection, Rotation, and Reflection Shang-Hua Teng.
Lecture 20 SVD and Its Applications Shang-Hua Teng.
Lecture 18 Eigenvalue Problems II Shang-Hua Teng.
CS240A: Conjugate Gradients and the Model Problem.
Tutorial 10 Iterative Methods and Matrix Norms. 2 In an iterative process, the k+1 step is defined via: Iterative processes Eigenvector decomposition.
Linear Least Squares QR Factorization. Systems of linear equations Problem to solve: M x = b Given M x = b : Is there a solution? Is the solution unique?
ECE 552 Numerical Circuit Analysis Chapter Five RELAXATION OR ITERATIVE TECHNIQUES FOR THE SOLUTION OF LINEAR EQUATIONS Copyright © I. Hajj 2012 All rights.
CSE 245: Computer Aided Circuit Simulation and Verification
Dirac Notation and Spectral decomposition
1cs542g-term Notes  Extra class next week (Oct 12, not this Friday)  To submit your assignment: me the URL of a page containing (links to)
MATH 685/ CSI 700/ OR 682 Lecture Notes Lecture 6. Eigenvalue problems.
Stats & Linear Models.
5.1 Orthogonality.
9 1 Performance Optimization. 9 2 Basic Optimization Algorithm p k - Search Direction  k - Learning Rate or.
ECE 530 – Analysis Techniques for Large-Scale Electrical Systems Prof. Hao Zhu Dept. of Electrical and Computer Engineering University of Illinois at Urbana-Champaign.
Introduction The central problems of Linear Algebra are to study the properties of matrices and to investigate the solutions of systems of linear equations.
Algorithms for a large sparse nonlinear eigenvalue problem Yusaku Yamamoto Dept. of Computational Science & Engineering Nagoya University.
1 Iterative Solution Methods Starts with an initial approximation for the solution vector (x 0 ) At each iteration updates the x vector by using the sytem.
Linear Algebra (Aljabar Linier) Week 10 Universitas Multimedia Nusantara Serpong, Tangerang Dr. Ananda Kusuma
Linear algebra: matrix Eigen-value Problems
1 Unconstrained Optimization Objective: Find minimum of F(X) where X is a vector of design variables We may know lower and upper bounds for optimum No.
Vector Norms and the related Matrix Norms. Properties of a Vector Norm: Euclidean Vector Norm: Riemannian metric:
Programming assignment #2 Solving a parabolic PDE using finite differences Numerical Methods for PDEs Spring 2007 Jim E. Jones.
Orthogonalization via Deflation By Achiya Dax Hydrological Service Jerusalem, Israel
Elementary Linear Algebra Anton & Rorres, 9th Edition
A Note on Rectangular Quotients By Achiya Dax Hydrological Service Jerusalem, Israel
1 Marijn Bartel Schreuders Supervisor: Dr. Ir. M.B. Van Gijzen Date:Monday, 24 February 2014.
CSE 245: Computer Aided Circuit Simulation and Verification Matrix Computations: Iterative Methods I Chung-Kuan Cheng.
Algorithms 2005 Ramesh Hariharan. Algebraic Methods.
Large Timestep Issues Lecture 12 Alessandra Nardi Thanks to Prof. Sangiovanni, Prof. Newton, Prof. White, Deepak Ramaswamy, Michal Rewienski, and Karen.
Case Study in Computational Science & Engineering - Lecture 5 1 Iterative Solution of Linear Systems Jacobi Method while not converged do { }
Direct Methods for Sparse Linear Systems Lecture 4 Alessandra Nardi Thanks to Prof. Jacob White, Suvranu De, Deepak Ramaswamy, Michal Rewienski, and Karen.
Direct Methods for Linear Systems Lecture 3 Alessandra Nardi Thanks to Prof. Jacob White, Suvranu De, Deepak Ramaswamy, Michal Rewienski, and Karen Veroy.
Signal & Weight Vector Spaces
Solving Scalar Linear Systems A Little Theory For Jacobi Iteration
Network Systems Lab. Korea Advanced Institute of Science and Technology No.1 Maximum Norms & Nonnegative Matrices  Weighted maximum norm e.g.) x1x1 x2x2.
MA5233 Lecture 6 Krylov Subspaces and Conjugate Gradients Wayne M. Lawton Department of Mathematics National University of Singapore 2 Science Drive 2.
ECE 530 – Analysis Techniques for Large-Scale Electrical Systems
Krylov-Subspace Methods - II Lecture 7 Alessandra Nardi Thanks to Prof. Jacob White, Deepak Ramaswamy, Michal Rewienski, and Karen Veroy.
1 Objective To provide background material in support of topics in Digital Image Processing that are based on matrices and/or vectors. Review Matrices.
Conjugate gradient iteration One matrix-vector multiplication per iteration Two vector dot products per iteration Four n-vectors of working storage x 0.
ALGEBRAIC EIGEN VALUE PROBLEMS
The Landscape of Sparse Ax=b Solvers Direct A = LU Iterative y’ = Ay Non- symmetric Symmetric positive definite More RobustLess Storage More Robust More.
Tutorial 6. Eigenvalues & Eigenvectors Reminder: Eigenvectors A vector x invariant up to a scaling by λ to a multiplication by matrix A is called.
Lecture 11 Alessandra Nardi
Krylov-Subspace Methods - I
Direct Methods for Sparse Linear Systems
Solving Linear Systems Ax=b
Euclidean Inner Product on Rn
CSE 245: Computer Aided Circuit Simulation and Verification
CSE 245: Computer Aided Circuit Simulation and Verification
Conjugate Gradient Method
Symmetric Matrices and Quadratic Forms
Performance Optimization
Symmetric Matrices and Quadratic Forms
Presentation transcript:

Krylov-Subspace Methods - I Lecture 6 Alessandra Nardi Thanks to Prof. Jacob White, Deepak Ramaswamy, Michal Rewienski, and Karen Veroy

Last lecture review Iterative Methods Overview –Stationary –Non Stationary QR factorization to solve Mx=b –Modified Gram-Schmidt Algorithm –QR Pivoting –Minimization View of QR Basic Minimization approach Orthogonalized Search Directions Pointer to Krylov Subspace Methods

Last lecture reminder QR Factorization – By picture

For i = 1 to N “For each Target Column” For j = 1 to i-1 “For each Source Column left of target” end Normalize Orthogonalize Search Direction QR Factorization – Minimization View Minimization Algorithm

Iterative Methods Solve Mx=b minimizing the residual r=b-Mx Stationary: x (k+1) =Gx (k) +c Jacobi Gauss-Seidel Successive Overrelaxation Non Stationary: x (k+1) =x (k) +a k p k CG (Conjugate Gradient)  A symmetric and positive definite GCR (Generalized Conjugate Residual) GMRES, etc etc

Iterative Methods - CG Convergence is related to: –Number of distinct eigenvalues –Ratio between max and min eigenvalue Why ? How?

General Subspace Minimization Algorithm –Review orthogonalization and projection formulas Generalized Conjugate Residual Algorithm –Krylov-subspace –Simplification in the symmetric case. –Convergence properties Eigenvalue and Eigenvector Review –Norms and Spectral Radius –Spectral Mapping Theorem Outline

Arbitrary Subspace Methods Residual Minimization

Use Gram-Schmidt on Mw i’s ! Arbitrary Subspace Methods Residual Minimization

Arbitrary Subspace Methods Orthogonalization

Arbitrary Subspace Solution Algorithm 1.Given M, b and a set of search directions: {w 0,…,w k } 2.Make w i ’s MM T orthogonal and get new search directions: {p 0,…,p k } 3.Minimize the residual:

For i = 0 to k For j = 1 to i-1 end Normalize Orthogonalize Search Direction Update Solution Arbitrary Subspace Solution Algorithm

Krylov Subspace How about the initial set of search directions {w 0,…,w k } ? A particular choice that is commonly used is: {w 0,…,w k }  {b, Mb, M 2 b…}  m (A,v)  span{v, Av, A 2 v, …, A m-1 v} is called Krylov Subspace

kth order polynomial Krylov Subspace Methods

Krylov Subspace Methods Subspace Generation The set of residuals also can be used as a representation of the Krylov-Subspace Generalized Conjugate Residual Algorithm Nice because the residuals generate next search directions

Determine optimal stepsize in kth search direction Update the solution (trying to minimize residual) and the residual Compute the new orthogonalized search direction (by using the most recent residual) Krylov-Subspace Methods Generalized Conjugate Residual Method (k-th step)

Vector inner products, O(n) Matrix-vector product, O(n) if sparse Vector Adds, O(n) O(k) inner products, total cost O(nk) If M is sparse, as k (# of iters) approaches n, Better Converge Fast! Krylov-Subspace Methods Generalized Conjugate Residual Method (Computational Complexity for k-th step)

If k (# of iters )  n, then symmetric, sparse, GCR is O(n 2 ) Better Converge Fast! An Amazing fact that will not be derived Orthogonalization in one step Krylov-Subspace Methods Generalized Conjugate Residual Method (Symmetric Case – Conjugate Gradient Method)

Summary What is an iterative non stationary method: x (k+1) =x (k) +a k p k How search to calculate: –Search directions (p k ) –Step along search directions (a k ) Krylov Subspace  GCR GCR is O(k 2 n) –Better converge fast!  Now look at convergence properties of GCR

Krylov Methods Convergence Analysis Basic properties

Krylov Methods Convergence Analysis Optimality of GCR poly GCR optimality property (key property of the algorithm): GCR picks the best (k+1)-th order polynomial minimizing and subject to: 

GCR Optimality Property Therefore Any polynomial which satisfies the constraints can be used to get an upper bound on Krylov Methods Convergence Analysis Optimality of GCR poly

Eigenvalues and eigenvectors of a matrix M satisfy eigenvector eigenvalue Eigenvalues and eigenvectors review Basic definitions

Almost all NxN matrices have N linearly independent Eigenvectors The set of all eigenvalues of M is known as the Spectrum of M Eigenvalues and eigenvectors review A symplifying assumption

Almost all NxN matrices have N linearly independent Eigenvectors Eigenvalues and eigenvectors review A symplifying assumption

The spectral Radius of M is the radius of the smallest circle, centered at the origin, which encloses all of M’s eigenvalues Eigenvalues and eigenvectors review Spectral radius

L 2 (Euclidean) norm : L 1 norm : L  norm : Unit circle Unit square 1 1 Eigenvalues and eigenvectors review Vector norms

Vector induced norm : Induced norm of A is the maximum “magnification” of by = max abs column sum = max abs row sum = (largest eigenvalue of A T A) 1/2 Eigenvalues and eigenvectors review Matrix norms

Theorem: Any induced norm is a bound on the spectral radius Proof: Eigenvalues and eigenvectors review Induced norms

Given a polynomial Apply the polynomial to a matrix Then Useful Eigenproperties Spectral Mapping Theorem

Krylov Methods Convergence Analysis Overview where is any (k+1)-th order polynomial subject to:  may be used to get an upper bound on Matrix norm propertyGCR optimality property

Review on eigenvalues and eigenvectors –Induced norms: relate matrix eigenvalues to the matrix norms –Spectral mapping theorem: relate matrix eigenvalues to matrix polynomials Now ready to relate the convergence properties of Krylov Subspace methods to eigenvalues of M Krylov Methods Convergence Analysis Overview

Summary Generalized Conjugate Residual Algorithm –Krylov-subspace –Simplification in the symmetric case –Convergence properties Eigenvalue and Eigenvector Review –Norms and Spectral Radius –Spectral Mapping Theorem