Lecture Notes: Constraint Optimization

Size: px
Start display at page:

Download "Lecture Notes: Constraint Optimization"

Transcription

1 Lecture Notes: Constraint Optimization Gerhard Neumann January 6, Constraint Optimization Problems In constraint optimization we want to maximize a function f(x) under the constraints that g i (x) = 0. For simplicity, we will only consider equality constraints. We can formalize this problem as argmax x f(x) s.t.:g i (x i ) = 0, i. (1) We look for a point x that lies on the constraint surface and maximizes f. Lets first consider the simpler case with only one constraint g(x) = 0- The constraint can be approximated by a Taylor expansion around a point c, i.e., g(x + ɛ) g(x) + ɛ T g(x). We want the point x + ɛ to be also on the surface, it follows that ɛ T g(x) = 0. Stated differently, g(x) is normal to the constraint surface. From Figure 1, we can see that the gradient f(x ) needs to be orthogonal to the constraint surface, i.e., f(x ) and g(x ) are (anti-) parallel. Stated differently, there needs to be a factor λ such that f(x ) + λ g(x ) = 0 (2) 1.1 The Lagrangian Function Now we can introduce the Lagrangian function L(x, λ))f(x) + λg(x), where λ is a Lagrangian multiplier. function, i.e., We want to find a sattel point of this L(x, λ) = 0 x (3) L(x, λ) = 0 λ (4) Note that these two conditions reconstruct the condition from Equation 2 and the constraint g(x) = 0. 1

2 Figure 1: Illustration of the constraint surface, the gradient directions g(x) and f(x) of the constraint and objective respectively. Taken from [Bishop, 2006]. 1.2 The Lagrangian Dual Function Under certain conditions, it is easier to optimize the lagrangian dual function of an optimization problem. The dual function is given by h(λ) = max L(x, λ). x The dual function is only a function of the lagrangian multipliers λ and it is, per definition, h(λ) f(x) + λg(x). For an optimal point x, the condition simplifies to h(λ) f(x ), as g(x ) = 0. Since h(λ) f(x ), the optimal point for λ is obtained by minimizing the dual function λ = argmin λ h(λ). (5) Under certain conditions, which are called Karesh-Kuhn-Tucker KKT conditions [Wikipedia, a], optimizing the dual is equivalent to solving the primal optimization problem, i.e., λ can be used to obtain x by x = argmax x L(x, λ ). (6) The question is why should we optimize the dual instead of the primal? Often, the dual is easier to solve. The number of variables is equivalent to the number of constraints, which is typically less then the dimensionality of x. We can also show that h(λ) is convex, even if f(x) is not. Some more notes on the dual: If the primal is maximized, the dual is minimized and vice versa. 2

3 Inequality constraints can be handled almost equivalent to equality constraints, with the exception that every inequality constraint in the primal problem also adds a inequality constraint to the dual, where the constraints are put on the lagrangian multipliers that are used for the inequality constraints. There are many differnt regularity (KKT) conditions, such that finding an optimal solution of the dual can be used for an optimal solution of the primal. The simplest one which we can use for most cases is the Slater condition [Wikipedia, b]. 1.3 Cookbook for constraint optimization We consider the following problem with equality and inequality constraints argmax x f(x) s.t.: g i (x i ) = 0, i, c j (x) 0, j (7) We need to consider the following steps in order to solve this constraint optimization problem: 1. Check for KKT conditions 2. Write down the Lagrangian L(x, λ, η) = f(x) + i λ i g(x) + j η j c j (x). The parametes η j are again lagrangian multipliers for the inequality constraints. 3. Maximize Lagrangian for x, i.e. x = l(λ, η) = argmax x L(x, λ, η). Note that x is a function of the lagrangian multipliers λ and η. Hence, this step is typically only feasible if we can obtain x analytically. 4. Set x back into the Lagrangian to obtain the dualfunction h(λ, η) = L(l(λ, η), λ, η). 5. Solve the dual optimization problem 6. Obtain optimal solution by 2 Examples [λ, η ] = argmin h(λ, η), s.t.: η i 0, i x = l(λ, η ). In this section, we will quickly review several examples that are important in the context of robot learning. 3

4 2.1 Resolved Velocity Control aka. Jacobian Pseudo Inverse We want to solve the following problem argmin q 1 2 qt q, s.t.: ẋ = J q. This is a convex problem, so the KKT condition is satisfied. The Lagrangian of this optimization problem is given by L( q, λ) = 1 2 qt q + λ T (ẋ J q). We obtain the optimal solution for q as a function of λ by setting the derivative of the Lagrangian to 0 and solving for q which yields L( q, λ) q = q T λ T J = 0, q T = l(λ) = λ T J. Setting the previous equation back into the Lagrangian results in the dual function h(λ) = 1 2 λt JJ T λ + λ T (ẋ JJ T λ) = λ T ẋ 1 2 λt JJ T λ. To solve the dual optimization problem, we have to find λ = argmax λ h(λ). We do this again by setting the derivative of h(λ) to zero, i.e., which yields h(λ) λ = ẋt λ T JJ T = 0, λ T = ẋ(jj T ) 1. Setting λ T back into the optimal solution l(λ), we obtain the solution q T = ẋ T (JJ T ) 1 J q = J T (JJ T ) 1 ẋ, which corresponds to the solution we know from the lecture. 2.2 Gradient Descent with Pre-Defined Metric Gradient Descent with a pre-defined metric we want to find the update direction θ for a parameter vector θ which is most similar to the standard gradient g = f(θ) T θ, where f(θ) is the function we want to optimize, which has a limited metric L M ( θ) = θm θ ɛ, where M is a positive definite, symmetric matrix. Hence, the optimization problem is defined by θ = argmax θ θ T g s.t.: θ T M θ ɛ 4

5 The objective as well as the constraint are convex, hence, the KKT conditions are satisfied. The Lagrangian of the optimization problem is given by L( θ, λ) = θ T g + λ(ɛ θ T M θ). We obtain the otpimal solution for θ by setting the derivative of the Lagrangian to 0, L( θ, λ) = g T 2λ θ T M = 0 T, θ which yields θ = M 1 g M 1 g. 2λ Note that we can always invert M as we assumed that M is positive definite. The previous equation already reveals a basic result for the gradient update. The update is always propertional to the standard gradient, which is transformed by multiplying the inverse of the metric matrix M. We still have to compute the learning rate λ which is a lagrangian multiplier. The multiplier is again obtained by optimizing the dual function. The dual function is given by h(λ) = gm 1 g 2λ + λɛ gt M 1 MM 1 g 4λ = λɛ + gt M 1 g. 4λ We optimize the dual function again by setting its derivative to 0. Rearranging terms yields h(λ) λ = ɛ gt M 1 g 4λ 2 = 0. λ 2 = gt M 1 g. 4ɛ As the lagrangian multiplier λ is due to the inequality constraint restricted to be positive, we have a unique solution at g λ = T M 1 g. 4ɛ Hence, we can also obtain the learning rate for the specified metric in closed form. Note that the learning rate λ depends on the bound ɛ, which also intuitively has of course to be the case. This derivation is the basis for all natural gradient algorithms, where M is typically the Fisher information matrix. 2.3 Relative Entropy Policy Search TODO 5

6 References [Bishop, 2006] Bishop, C. M. (2006). Pattern Recognition and Machine Learning (Information Science and Statistics). Springer-Verlag New York, Inc., Secaucus, NJ, USA. [Wikipedia, a] Wikipedia. Karush kuhn tucker conditions. [Wikipedia, b] Wikipedia. Slater condition. 6

Demo 1: KKT conditions with inequality constraints

Demo 1: KKT conditions with inequality constraints MS-C5 Introduction to Optimization Solutions 9 Ehtamo Demo : KKT conditions with inequality constraints Using the Karush-Kuhn-Tucker conditions, see if the points x (x, x ) (, 4) or x (x, x ) (6, ) are

More information

Introduction to Optimization

Introduction to Optimization Introduction to Optimization Constrained Optimization Marc Toussaint U Stuttgart Constrained Optimization General constrained optimization problem: Let R n, f : R n R, g : R n R m, h : R n R l find min

More information

Programming, numerics and optimization

Programming, numerics and optimization Programming, numerics and optimization Lecture C-4: Constrained optimization Łukasz Jankowski ljank@ippt.pan.pl Institute of Fundamental Technological Research Room 4.32, Phone +22.8261281 ext. 428 June

More information

Lecture 10: SVM Lecture Overview Support Vector Machines The binary classification problem

Lecture 10: SVM Lecture Overview Support Vector Machines The binary classification problem Computational Learning Theory Fall Semester, 2012/13 Lecture 10: SVM Lecturer: Yishay Mansour Scribe: Gitit Kehat, Yogev Vaknin and Ezra Levin 1 10.1 Lecture Overview In this lecture we present in detail

More information

Introduction to Machine Learning

Introduction to Machine Learning Introduction to Machine Learning Maximum Margin Methods Varun Chandola Computer Science & Engineering State University of New York at Buffalo Buffalo, NY, USA chandola@buffalo.edu Chandola@UB CSE 474/574

More information

ISM206 Lecture, April 26, 2005 Optimization of Nonlinear Objectives, with Non-Linear Constraints

ISM206 Lecture, April 26, 2005 Optimization of Nonlinear Objectives, with Non-Linear Constraints ISM206 Lecture, April 26, 2005 Optimization of Nonlinear Objectives, with Non-Linear Constraints Instructor: Kevin Ross Scribe: Pritam Roy May 0, 2005 Outline of topics for the lecture We will discuss

More information

Introduction to Constrained Optimization

Introduction to Constrained Optimization Introduction to Constrained Optimization Duality and KKT Conditions Pratik Shah {pratik.shah [at] lnmiit.ac.in} The LNM Institute of Information Technology www.lnmiit.ac.in February 13, 2013 LNMIIT MLPR

More information

Lecture 7: Support Vector Machine

Lecture 7: Support Vector Machine Lecture 7: Support Vector Machine Hien Van Nguyen University of Houston 9/28/2017 Separating hyperplane Red and green dots can be separated by a separating hyperplane Two classes are separable, i.e., each

More information

Shiqian Ma, MAT-258A: Numerical Optimization 1. Chapter 2. Convex Optimization

Shiqian Ma, MAT-258A: Numerical Optimization 1. Chapter 2. Convex Optimization Shiqian Ma, MAT-258A: Numerical Optimization 1 Chapter 2 Convex Optimization Shiqian Ma, MAT-258A: Numerical Optimization 2 2.1. Convex Optimization General optimization problem: min f 0 (x) s.t., f i

More information

Inverse KKT Motion Optimization: A Newton Method to Efficiently Extract Task Spaces and Cost Parameters from Demonstrations

Inverse KKT Motion Optimization: A Newton Method to Efficiently Extract Task Spaces and Cost Parameters from Demonstrations Inverse KKT Motion Optimization: A Newton Method to Efficiently Extract Task Spaces and Cost Parameters from Demonstrations Peter Englert Machine Learning and Robotics Lab Universität Stuttgart Germany

More information

9. Support Vector Machines. The linearly separable case: hard-margin SVMs. The linearly separable case: hard-margin SVMs. Learning objectives

9. Support Vector Machines. The linearly separable case: hard-margin SVMs. The linearly separable case: hard-margin SVMs. Learning objectives Foundations of Machine Learning École Centrale Paris Fall 25 9. Support Vector Machines Chloé-Agathe Azencot Centre for Computational Biology, Mines ParisTech Learning objectives chloe agathe.azencott@mines

More information

Computational Optimization. Constrained Optimization Algorithms

Computational Optimization. Constrained Optimization Algorithms Computational Optimization Constrained Optimization Algorithms Same basic algorithms Repeat Determine descent direction Determine step size Take a step Until Optimal But now must consider feasibility,

More information

Chapter 3 Numerical Methods

Chapter 3 Numerical Methods Chapter 3 Numerical Methods Part 1 3.1 Linearization and Optimization of Functions of Vectors 1 Problem Notation 2 Outline 3.1.1 Linearization 3.1.2 Optimization of Objective Functions 3.1.3 Constrained

More information

Theoretical Concepts of Machine Learning

Theoretical Concepts of Machine Learning Theoretical Concepts of Machine Learning Part 2 Institute of Bioinformatics Johannes Kepler University, Linz, Austria Outline 1 Introduction 2 Generalization Error 3 Maximum Likelihood 4 Noise Models 5

More information

Optimization III: Constrained Optimization

Optimization III: Constrained Optimization Optimization III: Constrained Optimization CS 205A: Mathematical Methods for Robotics, Vision, and Graphics Doug James (and Justin Solomon) CS 205A: Mathematical Methods Optimization III: Constrained Optimization

More information

Convex Optimization. Erick Delage, and Ashutosh Saxena. October 20, (a) (b) (c)

Convex Optimization. Erick Delage, and Ashutosh Saxena. October 20, (a) (b) (c) Convex Optimization (for CS229) Erick Delage, and Ashutosh Saxena October 20, 2006 1 Convex Sets Definition: A set G R n is convex if every pair of point (x, y) G, the segment beteen x and y is in A. More

More information

LECTURE 13: SOLUTION METHODS FOR CONSTRAINED OPTIMIZATION. 1. Primal approach 2. Penalty and barrier methods 3. Dual approach 4. Primal-dual approach

LECTURE 13: SOLUTION METHODS FOR CONSTRAINED OPTIMIZATION. 1. Primal approach 2. Penalty and barrier methods 3. Dual approach 4. Primal-dual approach LECTURE 13: SOLUTION METHODS FOR CONSTRAINED OPTIMIZATION 1. Primal approach 2. Penalty and barrier methods 3. Dual approach 4. Primal-dual approach Basic approaches I. Primal Approach - Feasible Direction

More information

Linear methods for supervised learning

Linear methods for supervised learning Linear methods for supervised learning LDA Logistic regression Naïve Bayes PLA Maximum margin hyperplanes Soft-margin hyperplanes Least squares resgression Ridge regression Nonlinear feature maps Sometimes

More information

Applied Lagrange Duality for Constrained Optimization

Applied Lagrange Duality for Constrained Optimization Applied Lagrange Duality for Constrained Optimization Robert M. Freund February 10, 2004 c 2004 Massachusetts Institute of Technology. 1 1 Overview The Practical Importance of Duality Review of Convexity

More information

A Short SVM (Support Vector Machine) Tutorial

A Short SVM (Support Vector Machine) Tutorial A Short SVM (Support Vector Machine) Tutorial j.p.lewis CGIT Lab / IMSC U. Southern California version 0.zz dec 004 This tutorial assumes you are familiar with linear algebra and equality-constrained optimization/lagrange

More information

Constrained Optimization COS 323

Constrained Optimization COS 323 Constrained Optimization COS 323 Last time Introduction to optimization objective function, variables, [constraints] 1-dimensional methods Golden section, discussion of error Newton s method Multi-dimensional

More information

A Truncated Newton Method in an Augmented Lagrangian Framework for Nonlinear Programming

A Truncated Newton Method in an Augmented Lagrangian Framework for Nonlinear Programming A Truncated Newton Method in an Augmented Lagrangian Framework for Nonlinear Programming Gianni Di Pillo (dipillo@dis.uniroma1.it) Giampaolo Liuzzi (liuzzi@iasi.cnr.it) Stefano Lucidi (lucidi@dis.uniroma1.it)

More information

Lec 11 Rate-Distortion Optimization (RDO) in Video Coding-I

Lec 11 Rate-Distortion Optimization (RDO) in Video Coding-I CS/EE 5590 / ENG 401 Special Topics (17804, 17815, 17803) Lec 11 Rate-Distortion Optimization (RDO) in Video Coding-I Zhu Li Course Web: http://l.web.umkc.edu/lizhu/teaching/2016sp.video-communication/main.html

More information

Lesson 11: Duality in linear programming

Lesson 11: Duality in linear programming Unit 1 Lesson 11: Duality in linear programming Learning objectives: Introduction to dual programming. Formulation of Dual Problem. Introduction For every LP formulation there exists another unique linear

More information

In other words, we want to find the domain points that yield the maximum or minimum values (extrema) of the function.

In other words, we want to find the domain points that yield the maximum or minimum values (extrema) of the function. 1 The Lagrange multipliers is a mathematical method for performing constrained optimization of differentiable functions. Recall unconstrained optimization of differentiable functions, in which we want

More information

Characterizing Improving Directions Unconstrained Optimization

Characterizing Improving Directions Unconstrained Optimization Final Review IE417 In the Beginning... In the beginning, Weierstrass's theorem said that a continuous function achieves a minimum on a compact set. Using this, we showed that for a convex set S and y not

More information

Convex Optimization and Machine Learning

Convex Optimization and Machine Learning Convex Optimization and Machine Learning Mengliu Zhao Machine Learning Reading Group School of Computing Science Simon Fraser University March 12, 2014 Mengliu Zhao SFU-MLRG March 12, 2014 1 / 25 Introduction

More information

QEM Optimization, WS 2017/18 Part 4. Constrained optimization

QEM Optimization, WS 2017/18 Part 4. Constrained optimization QEM Optimization, WS 2017/18 Part 4 Constrained optimization (about 4 Lectures) Supporting Literature: Angel de la Fuente, Mathematical Methods and Models for Economists, Chapter 7 Contents 4 Constrained

More information

CS545 Contents IX. Inverse Kinematics. Reading Assignment for Next Class. Analytical Methods Iterative (Differential) Methods

CS545 Contents IX. Inverse Kinematics. Reading Assignment for Next Class. Analytical Methods Iterative (Differential) Methods CS545 Contents IX Inverse Kinematics Analytical Methods Iterative (Differential) Methods Geometric and Analytical Jacobian Jacobian Transpose Method Pseudo-Inverse Pseudo-Inverse with Optimization Extended

More information

Introduction to Optimization

Introduction to Optimization Introduction to Optimization Second Order Optimization Methods Marc Toussaint U Stuttgart Planned Outline Gradient-based optimization (1st order methods) plain grad., steepest descent, conjugate grad.,

More information

Solution Methods Numerical Algorithms

Solution Methods Numerical Algorithms Solution Methods Numerical Algorithms Evelien van der Hurk DTU Managment Engineering Class Exercises From Last Time 2 DTU Management Engineering 42111: Static and Dynamic Optimization (6) 09/10/2017 Class

More information

Lecture 18: March 23

Lecture 18: March 23 0-725/36-725: Convex Optimization Spring 205 Lecturer: Ryan Tibshirani Lecture 8: March 23 Scribes: James Duyck Note: LaTeX template courtesy of UC Berkeley EECS dept. Disclaimer: These notes have not

More information

PRIMAL-DUAL INTERIOR POINT METHOD FOR LINEAR PROGRAMMING. 1. Introduction

PRIMAL-DUAL INTERIOR POINT METHOD FOR LINEAR PROGRAMMING. 1. Introduction PRIMAL-DUAL INTERIOR POINT METHOD FOR LINEAR PROGRAMMING KELLER VANDEBOGERT AND CHARLES LANNING 1. Introduction Interior point methods are, put simply, a technique of optimization where, given a problem

More information

Kernel Methods & Support Vector Machines

Kernel Methods & Support Vector Machines & Support Vector Machines & Support Vector Machines Arvind Visvanathan CSCE 970 Pattern Recognition 1 & Support Vector Machines Question? Draw a single line to separate two classes? 2 & Support Vector

More information

Convex Programs. COMPSCI 371D Machine Learning. COMPSCI 371D Machine Learning Convex Programs 1 / 21

Convex Programs. COMPSCI 371D Machine Learning. COMPSCI 371D Machine Learning Convex Programs 1 / 21 Convex Programs COMPSCI 371D Machine Learning COMPSCI 371D Machine Learning Convex Programs 1 / 21 Logistic Regression! Support Vector Machines Support Vector Machines (SVMs) and Convex Programs SVMs are

More information

Lecture 2 September 3

Lecture 2 September 3 EE 381V: Large Scale Optimization Fall 2012 Lecture 2 September 3 Lecturer: Caramanis & Sanghavi Scribe: Hongbo Si, Qiaoyang Ye 2.1 Overview of the last Lecture The focus of the last lecture was to give

More information

FMA901F: Machine Learning Lecture 3: Linear Models for Regression. Cristian Sminchisescu

FMA901F: Machine Learning Lecture 3: Linear Models for Regression. Cristian Sminchisescu FMA901F: Machine Learning Lecture 3: Linear Models for Regression Cristian Sminchisescu Machine Learning: Frequentist vs. Bayesian In the frequentist setting, we seek a fixed parameter (vector), with value(s)

More information

Optimization Methods: Optimization using Calculus Kuhn-Tucker Conditions 1. Module - 2 Lecture Notes 5. Kuhn-Tucker Conditions

Optimization Methods: Optimization using Calculus Kuhn-Tucker Conditions 1. Module - 2 Lecture Notes 5. Kuhn-Tucker Conditions Optimization Methods: Optimization using Calculus Kuhn-Tucker Conditions Module - Lecture Notes 5 Kuhn-Tucker Conditions Introduction In the previous lecture the optimization of functions of multiple variables

More information

Lecture 2 Optimization with equality constraints

Lecture 2 Optimization with equality constraints Lecture 2 Optimization with equality constraints Constrained optimization The idea of constrained optimisation is that the choice of one variable often affects the amount of another variable that can be

More information

CONLIN & MMA solvers. Pierre DUYSINX LTAS Automotive Engineering Academic year

CONLIN & MMA solvers. Pierre DUYSINX LTAS Automotive Engineering Academic year CONLIN & MMA solvers Pierre DUYSINX LTAS Automotive Engineering Academic year 2018-2019 1 CONLIN METHOD 2 LAY-OUT CONLIN SUBPROBLEMS DUAL METHOD APPROACH FOR CONLIN SUBPROBLEMS SEQUENTIAL QUADRATIC PROGRAMMING

More information

Contents. I Basics 1. Copyright by SIAM. Unauthorized reproduction of this article is prohibited.

Contents. I Basics 1. Copyright by SIAM. Unauthorized reproduction of this article is prohibited. page v Preface xiii I Basics 1 1 Optimization Models 3 1.1 Introduction... 3 1.2 Optimization: An Informal Introduction... 4 1.3 Linear Equations... 7 1.4 Linear Optimization... 10 Exercises... 12 1.5

More information

Sparse Optimization Lecture: Proximal Operator/Algorithm and Lagrange Dual

Sparse Optimization Lecture: Proximal Operator/Algorithm and Lagrange Dual Sparse Optimization Lecture: Proximal Operator/Algorithm and Lagrange Dual Instructor: Wotao Yin July 2013 online discussions on piazza.com Those who complete this lecture will know learn the proximal

More information

Constrained Optimization and Lagrange Multipliers

Constrained Optimization and Lagrange Multipliers Constrained Optimization and Lagrange Multipliers MATH 311, Calculus III J. Robert Buchanan Department of Mathematics Fall 2011 Constrained Optimization In the previous section we found the local or absolute

More information

Unconstrained Optimization Principles of Unconstrained Optimization Search Methods

Unconstrained Optimization Principles of Unconstrained Optimization Search Methods 1 Nonlinear Programming Types of Nonlinear Programs (NLP) Convexity and Convex Programs NLP Solutions Unconstrained Optimization Principles of Unconstrained Optimization Search Methods Constrained Optimization

More information

Convexity Theory and Gradient Methods

Convexity Theory and Gradient Methods Convexity Theory and Gradient Methods Angelia Nedić angelia@illinois.edu ISE Department and Coordinated Science Laboratory University of Illinois at Urbana-Champaign Outline Convex Functions Optimality

More information

Lecture 2 - Introduction to Polytopes

Lecture 2 - Introduction to Polytopes Lecture 2 - Introduction to Polytopes Optimization and Approximation - ENS M1 Nicolas Bousquet 1 Reminder of Linear Algebra definitions Let x 1,..., x m be points in R n and λ 1,..., λ m be real numbers.

More information

21-256: Lagrange multipliers

21-256: Lagrange multipliers 21-256: Lagrange multipliers Clive Newstead, Thursday 12th June 2014 Lagrange multipliers give us a means of optimizing multivariate functions subject to a number of constraints on their variables. Problems

More information

Machine Learning for Signal Processing Lecture 4: Optimization

Machine Learning for Signal Processing Lecture 4: Optimization Machine Learning for Signal Processing Lecture 4: Optimization 13 Sep 2015 Instructor: Bhiksha Raj (slides largely by Najim Dehak, JHU) 11-755/18-797 1 Index 1. The problem of optimization 2. Direct optimization

More information

Math 5593 Linear Programming Lecture Notes

Math 5593 Linear Programming Lecture Notes Math 5593 Linear Programming Lecture Notes Unit II: Theory & Foundations (Convex Analysis) University of Colorado Denver, Fall 2013 Topics 1 Convex Sets 1 1.1 Basic Properties (Luenberger-Ye Appendix B.1).........................

More information

Lecture 15: Log Barrier Method

Lecture 15: Log Barrier Method 10-725/36-725: Convex Optimization Spring 2015 Lecturer: Ryan Tibshirani Lecture 15: Log Barrier Method Scribes: Pradeep Dasigi, Mohammad Gowayyed Note: LaTeX template courtesy of UC Berkeley EECS dept.

More information

Linear Programming Problems

Linear Programming Problems Linear Programming Problems Two common formulations of linear programming (LP) problems are: min Subject to: 1,,, 1,2,,;, max Subject to: 1,,, 1,2,,;, Linear Programming Problems The standard LP problem

More information

California Institute of Technology Crash-Course on Convex Optimization Fall Ec 133 Guilherme Freitas

California Institute of Technology Crash-Course on Convex Optimization Fall Ec 133 Guilherme Freitas California Institute of Technology HSS Division Crash-Course on Convex Optimization Fall 2011-12 Ec 133 Guilherme Freitas In this text, we will study the following basic problem: maximize x C f(x) subject

More information

Sequential Coordinate-wise Algorithm for Non-negative Least Squares Problem

Sequential Coordinate-wise Algorithm for Non-negative Least Squares Problem CENTER FOR MACHINE PERCEPTION CZECH TECHNICAL UNIVERSITY Sequential Coordinate-wise Algorithm for Non-negative Least Squares Problem Woring document of the EU project COSPAL IST-004176 Vojtěch Franc, Miro

More information

Nonlinear Programming

Nonlinear Programming Nonlinear Programming SECOND EDITION Dimitri P. Bertsekas Massachusetts Institute of Technology WWW site for book Information and Orders http://world.std.com/~athenasc/index.html Athena Scientific, Belmont,

More information

Lecture 19: Convex Non-Smooth Optimization. April 2, 2007

Lecture 19: Convex Non-Smooth Optimization. April 2, 2007 : Convex Non-Smooth Optimization April 2, 2007 Outline Lecture 19 Convex non-smooth problems Examples Subgradients and subdifferentials Subgradient properties Operations with subgradients and subdifferentials

More information

Lagrange multipliers. Contents. Introduction. From Wikipedia, the free encyclopedia

Lagrange multipliers. Contents. Introduction. From Wikipedia, the free encyclopedia Lagrange multipliers From Wikipedia, the free encyclopedia In mathematical optimization problems, Lagrange multipliers, named after Joseph Louis Lagrange, is a method for finding the local extrema of a

More information

Kernels and Constrained Optimization

Kernels and Constrained Optimization Machine Learning 1 WS2014 Module IN2064 Sheet 8 Page 1 Machine Learning Worksheet 8 Kernels and Constrained Optimization 1 Kernelized k-nearest neighbours To classify the point x the k-nearest neighbours

More information

Department of Mathematics Oleg Burdakov of 30 October Consider the following linear programming problem (LP):

Department of Mathematics Oleg Burdakov of 30 October Consider the following linear programming problem (LP): Linköping University Optimization TAOP3(0) Department of Mathematics Examination Oleg Burdakov of 30 October 03 Assignment Consider the following linear programming problem (LP): max z = x + x s.t. x x

More information

Optimal Control Techniques for Dynamic Walking

Optimal Control Techniques for Dynamic Walking Optimal Control Techniques for Dynamic Walking Optimization in Robotics & Biomechanics IWR, University of Heidelberg Presentation partly based on slides by Sebastian Sager, Moritz Diehl and Peter Riede

More information

Augmented Lagrangian Methods

Augmented Lagrangian Methods Augmented Lagrangian Methods Mário A. T. Figueiredo 1 and Stephen J. Wright 2 1 Instituto de Telecomunicações, Instituto Superior Técnico, Lisboa, Portugal 2 Computer Sciences Department, University of

More information

Local and Global Minimum

Local and Global Minimum Local and Global Minimum Stationary Point. From elementary calculus, a single variable function has a stationary point at if the derivative vanishes at, i.e., 0. Graphically, the slope of the function

More information

A Brief Look at Optimization

A Brief Look at Optimization A Brief Look at Optimization CSC 412/2506 Tutorial David Madras January 18, 2018 Slides adapted from last year s version Overview Introduction Classes of optimization problems Linear programming Steepest

More information

Classification: Linear Discriminant Functions

Classification: Linear Discriminant Functions Classification: Linear Discriminant Functions CE-725: Statistical Pattern Recognition Sharif University of Technology Spring 2013 Soleymani Outline Discriminant functions Linear Discriminant functions

More information

Section 5 Convex Optimisation 1. W. Dai (IC) EE4.66 Data Proc. Convex Optimisation page 5-1

Section 5 Convex Optimisation 1. W. Dai (IC) EE4.66 Data Proc. Convex Optimisation page 5-1 Section 5 Convex Optimisation 1 W. Dai (IC) EE4.66 Data Proc. Convex Optimisation 1 2018 page 5-1 Convex Combination Denition 5.1 A convex combination is a linear combination of points where all coecients

More information

OPTIMIZATION METHODS

OPTIMIZATION METHODS D. Nagesh Kumar Associate Professor Department of Civil Engineering, Indian Institute of Science, Bangalore - 50 0 Email : nagesh@civil.iisc.ernet.in URL: http://www.civil.iisc.ernet.in/~nagesh Brief Contents

More information

Least-Squares Minimization Under Constraints EPFL Technical Report #

Least-Squares Minimization Under Constraints EPFL Technical Report # Least-Squares Minimization Under Constraints EPFL Technical Report # 150790 P. Fua A. Varol R. Urtasun M. Salzmann IC-CVLab, EPFL IC-CVLab, EPFL TTI Chicago TTI Chicago Unconstrained Least-Squares minimization

More information

Support Vector Machines.

Support Vector Machines. Support Vector Machines srihari@buffalo.edu SVM Discussion Overview 1. Overview of SVMs 2. Margin Geometry 3. SVM Optimization 4. Overlapping Distributions 5. Relationship to Logistic Regression 6. Dealing

More information

Data Mining Chapter 8: Search and Optimization Methods Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University

Data Mining Chapter 8: Search and Optimization Methods Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University Data Mining Chapter 8: Search and Optimization Methods Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University Search & Optimization Search and Optimization method deals with

More information

Using Analytic QP and Sparseness to Speed Training of Support Vector Machines

Using Analytic QP and Sparseness to Speed Training of Support Vector Machines Using Analytic QP and Sparseness to Speed Training of Support Vector Machines John C. Platt Microsoft Research 1 Microsoft Way Redmond, WA 9805 jplatt@microsoft.com Abstract Training a Support Vector Machine

More information

Parallel and Distributed Graph Cuts by Dual Decomposition

Parallel and Distributed Graph Cuts by Dual Decomposition Parallel and Distributed Graph Cuts by Dual Decomposition Presented by Varun Gulshan 06 July, 2010 What is Dual Decomposition? What is Dual Decomposition? It is a technique for optimization: usually used

More information

for Approximating the Analytic Center of a Polytope Abstract. The analytic center of a polytope P + = fx 0 : Ax = b e T x =1g log x j

for Approximating the Analytic Center of a Polytope Abstract. The analytic center of a polytope P + = fx 0 : Ax = b e T x =1g log x j A Lagrangian Relaxation Method for Approximating the Analytic Center of a Polytope Masakazu Kojima y Nimrod Megiddo z Shinji Mizuno x September 1992 Abstract. The analytic center of a polytope P + = fx

More information

Department of Electrical and Computer Engineering

Department of Electrical and Computer Engineering LAGRANGIAN RELAXATION FOR GATE IMPLEMENTATION SELECTION Yi-Le Huang, Jiang Hu and Weiping Shi Department of Electrical and Computer Engineering Texas A&M University OUTLINE Introduction and motivation

More information

Division of the Humanities and Social Sciences. Convex Analysis and Economic Theory Winter Separation theorems

Division of the Humanities and Social Sciences. Convex Analysis and Economic Theory Winter Separation theorems Division of the Humanities and Social Sciences Ec 181 KC Border Convex Analysis and Economic Theory Winter 2018 Topic 8: Separation theorems 8.1 Hyperplanes and half spaces Recall that a hyperplane in

More information

EC5555 Economics Masters Refresher Course in Mathematics September Lecture 6 Optimization with equality constraints Francesco Feri

EC5555 Economics Masters Refresher Course in Mathematics September Lecture 6 Optimization with equality constraints Francesco Feri EC5555 Economics Masters Refresher Course in Mathematics September 2013 Lecture 6 Optimization with equality constraints Francesco Feri Constrained optimization The idea of constrained optimisation is

More information

Search direction improvement for gradient-based optimization problems

Search direction improvement for gradient-based optimization problems Computer Aided Optimum Design in Engineering IX 3 Search direction improvement for gradient-based optimization problems S Ganguly & W L Neu Aerospace and Ocean Engineering, Virginia Tech, USA Abstract

More information

AM 221: Advanced Optimization Spring 2016

AM 221: Advanced Optimization Spring 2016 AM 221: Advanced Optimization Spring 2016 Prof. Yaron Singer Lecture 2 Wednesday, January 27th 1 Overview In our previous lecture we discussed several applications of optimization, introduced basic terminology,

More information

Lecture 5: Properties of convex sets

Lecture 5: Properties of convex sets Lecture 5: Properties of convex sets Rajat Mittal IIT Kanpur This week we will see properties of convex sets. These properties make convex sets special and are the reason why convex optimization problems

More information

Using Analytic QP and Sparseness to Speed Training of Support Vector Machines

Using Analytic QP and Sparseness to Speed Training of Support Vector Machines Using Analytic QP and Sparseness to Speed Training of Support Vector Machines John C. Platt Microsoft Research 1 Microsoft Way Redmond, WA 98052 jplatt@microsoft.com Abstract Training a Support Vector

More information

COMPUTATIONAL INTELLIGENCE (INTRODUCTION TO MACHINE LEARNING) SS18. Lecture 2: Linear Regression Gradient Descent Non-linear basis functions

COMPUTATIONAL INTELLIGENCE (INTRODUCTION TO MACHINE LEARNING) SS18. Lecture 2: Linear Regression Gradient Descent Non-linear basis functions COMPUTATIONAL INTELLIGENCE (INTRODUCTION TO MACHINE LEARNING) SS18 Lecture 2: Linear Regression Gradient Descent Non-linear basis functions LINEAR REGRESSION MOTIVATION Why Linear Regression? Simplest

More information

MATH3016: OPTIMIZATION

MATH3016: OPTIMIZATION MATH3016: OPTIMIZATION Lecturer: Dr Huifu Xu School of Mathematics University of Southampton Highfield SO17 1BJ Southampton Email: h.xu@soton.ac.uk 1 Introduction What is optimization? Optimization is

More information

Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers

Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers. Lagrange Multipliers In this section we present Lagrange s method for maximizing or minimizing a general function f(x, y, z) subject to a constraint (or side condition) of the form g(x, y, z) = k. Figure 1 shows this curve

More information

Gate Sizing by Lagrangian Relaxation Revisited

Gate Sizing by Lagrangian Relaxation Revisited Gate Sizing by Lagrangian Relaxation Revisited Jia Wang, Debasish Das, and Hai Zhou Electrical Engineering and Computer Science Northwestern University Evanston, Illinois, United States October 17, 2007

More information

Virtual Manikin Controller Calculating the movement of a human model Master s thesis in Complex Adaptive Systems DANIEL GLEESON

Virtual Manikin Controller Calculating the movement of a human model Master s thesis in Complex Adaptive Systems DANIEL GLEESON Virtual Manikin Controller Calculating the movement of a human model Master s thesis in Complex Adaptive Systems DANIEL GLEESON Department of Applied Mechanics Division of Vehicle Engineering and Autonomous

More information

Convex Optimization Lecture 2

Convex Optimization Lecture 2 Convex Optimization Lecture 2 Today: Convex Analysis Center-of-mass Algorithm 1 Convex Analysis Convex Sets Definition: A set C R n is convex if for all x, y C and all 0 λ 1, λx + (1 λ)y C Operations that

More information

Data Mining: Concepts and Techniques. Chapter 9 Classification: Support Vector Machines. Support Vector Machines (SVMs)

Data Mining: Concepts and Techniques. Chapter 9 Classification: Support Vector Machines. Support Vector Machines (SVMs) Data Mining: Concepts and Techniques Chapter 9 Classification: Support Vector Machines 1 Support Vector Machines (SVMs) SVMs are a set of related supervised learning methods used for classification Based

More information

Problem Set #3 (With Corrections)

Problem Set #3 (With Corrections) IC-32 Optimal Control Winter 2006 Benoît Chachuat ME C2 401, Ph: 33844, benoit.chachuat@epfl.ch Problem Set #3 (With Corrections) 1. Consider the following NLP problem: min x IR 3 f(x) := x2 1 + x 1 x

More information

Intoduction to Optimization. Computer Graphics CMU /15-662, Fall 2016

Intoduction to Optimization. Computer Graphics CMU /15-662, Fall 2016 Intoduction to Optimization Computer Graphics CMU 15-462/15-662, Fall 2016 Last time: Rigging, FK, IK Last Class: - Rigging, FK, IK - Animate high-resolution meshes using a much smaller number of parameters

More information

6.854 Advanced Algorithms. Scribes: Jay Kumar Sundararajan. Duality

6.854 Advanced Algorithms. Scribes: Jay Kumar Sundararajan. Duality 6.854 Advanced Algorithms Scribes: Jay Kumar Sundararajan Lecturer: David Karger Duality This lecture covers weak and strong duality, and also explains the rules for finding the dual of a linear program,

More information

Programs. Introduction

Programs. Introduction 16 Interior Point I: Linear Programs Lab Objective: For decades after its invention, the Simplex algorithm was the only competitive method for linear programming. The past 30 years, however, have seen

More information

David G. Luenberger Yinyu Ye. Linear and Nonlinear. Programming. Fourth Edition. ö Springer

David G. Luenberger Yinyu Ye. Linear and Nonlinear. Programming. Fourth Edition. ö Springer David G. Luenberger Yinyu Ye Linear and Nonlinear Programming Fourth Edition ö Springer Contents 1 Introduction 1 1.1 Optimization 1 1.2 Types of Problems 2 1.3 Size of Problems 5 1.4 Iterative Algorithms

More information

A primal-dual framework for mixtures of regularizers

A primal-dual framework for mixtures of regularizers A primal-dual framework for mixtures of regularizers Baran Gözcü baran.goezcue@epfl.ch Laboratory for Information and Inference Systems (LIONS) École Polytechnique Fédérale de Lausanne (EPFL) Switzerland

More information

All lecture slides will be available at CSC2515_Winter15.html

All lecture slides will be available at  CSC2515_Winter15.html CSC2515 Fall 2015 Introduc3on to Machine Learning Lecture 9: Support Vector Machines All lecture slides will be available at http://www.cs.toronto.edu/~urtasun/courses/csc2515/ CSC2515_Winter15.html Many

More information

Week 5. Convex Optimization

Week 5. Convex Optimization Week 5. Convex Optimization Lecturer: Prof. Santosh Vempala Scribe: Xin Wang, Zihao Li Feb. 9 and, 206 Week 5. Convex Optimization. The convex optimization formulation A general optimization problem is

More information

CS 473: Algorithms. Ruta Mehta. Spring University of Illinois, Urbana-Champaign. Ruta (UIUC) CS473 1 Spring / 36

CS 473: Algorithms. Ruta Mehta. Spring University of Illinois, Urbana-Champaign. Ruta (UIUC) CS473 1 Spring / 36 CS 473: Algorithms Ruta Mehta University of Illinois, Urbana-Champaign Spring 2018 Ruta (UIUC) CS473 1 Spring 2018 1 / 36 CS 473: Algorithms, Spring 2018 LP Duality Lecture 20 April 3, 2018 Some of the

More information

Convex Optimization. Lijun Zhang Modification of

Convex Optimization. Lijun Zhang   Modification of Convex Optimization Lijun Zhang zlj@nju.edu.cn http://cs.nju.edu.cn/zlj Modification of http://stanford.edu/~boyd/cvxbook/bv_cvxslides.pdf Outline Introduction Convex Sets & Functions Convex Optimization

More information

Outline. CS38 Introduction to Algorithms. Linear programming 5/21/2014. Linear programming. Lecture 15 May 20, 2014

Outline. CS38 Introduction to Algorithms. Linear programming 5/21/2014. Linear programming. Lecture 15 May 20, 2014 5/2/24 Outline CS38 Introduction to Algorithms Lecture 5 May 2, 24 Linear programming simplex algorithm LP duality ellipsoid algorithm * slides from Kevin Wayne May 2, 24 CS38 Lecture 5 May 2, 24 CS38

More information

minimise f(x) subject to g(x) = b, x X. inf f(x) = inf L(x,

minimise f(x) subject to g(x) = b, x X. inf f(x) = inf L(x, Optimisation Lecture 3 - Easter 2017 Michael Tehranchi Lagrangian necessity Consider the problem Let minimise f(x) subject to g(x) = b, x X. L(x, λ) = f(x) + λ (b g(x)) be the Lagrangian. Notice that for

More information

Errata for the book Optimization: principles and algorithms

Errata for the book Optimization: principles and algorithms Errata for the book Optimization: principles and algorithms Michel Bierlaire August 16, 2018 1 Errors Page 16 Equation (1.31) is ambiguous. It should be where h(x) = g(f(x)). argmin x X R nf(x) = argmin

More information

MAC2313 Test 3 A E g(x, y, z) dy dx dz

MAC2313 Test 3 A E g(x, y, z) dy dx dz MAC2313 Test 3 A (5 pts) 1. If the function g(x, y, z) is integrated over the cylindrical solid bounded by x 2 + y 2 = 3, z = 1, and z = 7, the correct integral in Cartesian coordinates is given by: A.

More information

Augmented Lagrangian Methods

Augmented Lagrangian Methods Augmented Lagrangian Methods Stephen J. Wright 1 2 Computer Sciences Department, University of Wisconsin-Madison. IMA, August 2016 Stephen Wright (UW-Madison) Augmented Lagrangian IMA, August 2016 1 /

More information