1 Algorithms and Software for Large-Scale Nonlinear Optimization OTC day, 6 Nov 2003 Richard Waltz, Northwestern University Project I: Large-scale Active-Set.

Slides:



Advertisements
Similar presentations
Testing Linear Pricing Algorithms for use in Ascending Combinatorial Auctions (A5) Giro Cavallo David Johnson Emrah Kostem.
Advertisements

Zhen Lu CPACT University of Newcastle MDC Technology Reduced Hessian Sequential Quadratic Programming(SQP)
Survey of gradient based constrained optimization algorithms Select algorithms based on their popularity. Additional details and additional algorithms.
Engineering Optimization
Optimization in Engineering Design Georgia Institute of Technology Systems Realization Laboratory 123 “True” Constrained Minimization.
Optimization Methods TexPoint fonts used in EMF.
1 TTK4135 Optimization and control B.Foss Spring semester 2005 TTK4135 Optimization and control Spring semester 2005 Scope - this you shall learn Optimization.
Optimization of thermal processes2007/2008 Optimization of thermal processes Maciej Marek Czestochowa University of Technology Institute of Thermal Machinery.
Binh research 08 Dec 2009 Working on ST and non-convex time stepper within Bullet DONE with initial solver for frictionless ST. Working on non-convex time.
Convex Optimization Chapter 1 Introduction. What, Why and How  What is convex optimization  Why study convex optimization  How to study convex optimization.
Inexact SQP Methods for Equality Constrained Optimization Frank Edward Curtis Department of IE/MS, Northwestern University with Richard Byrd and Jorge.
PROCESS INTEGRATED DESIGN WITHIN A MODEL PREDICTIVE CONTROL FRAMEWORK Mario Francisco, Pastora Vega, Omar Pérez University of Salamanca – Spain University.
The Most Important Concept in Optimization (minimization)  A point is said to be an optimal solution of a unconstrained minimization if there exists no.
Using MPC in MPC Tim Robinson.
Charge-Sensitive TCP and Rate Control Richard J. La Department of EECS UC Berkeley November 22, 1999.
Engineering Optimization
1 A Second Stage Network Recourse Problem in Stochastic Airline Crew Scheduling Joyce W. Yen University of Michigan John R. Birge Northwestern University.
Reformulated - SVR as a Constrained Minimization Problem subject to n+1+2m variables and 2m constrains minimization problem Enlarge the problem size and.
Unconstrained Optimization Problem
MAE 552 – Heuristic Optimization Lecture 10 February 13, 2002.
U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Optimal Fixed-Size Controllers for Decentralized POMDPs Christopher Amato Daniel.
Lecture outline Support vector machines. Support Vector Machines Find a linear hyperplane (decision boundary) that will separate the data.
Interior-Point Solvers
Digital Filter Stepsize Control in DASPK and its Effect on Control Optimization Performance Kirsten Meeker University of California, Santa Barbara.
Optimization of Linear Problems: Linear Programming (LP) © 2011 Daniel Kirschen and University of Washington 1.
Computational Optimization
IE 607 Constrained Design: Using Constraints to Advantage in Adaptive Optimization in Manufacturing.
A New Method For Numerical Constrained Optimization Ronald N. Perry Mitsubishi Electric Research Laboratories.
1 Hybrid methods for solving large-scale parameter estimation problems Carlos A. Quintero 1 Miguel Argáez 1 Hector Klie 2 Leticia Velázquez 1 Mary Wheeler.
1 Chapter 8 Nonlinear Programming with Constraints.
Frank Edward Curtis Northwestern University Joint work with Richard Byrd and Jorge Nocedal February 12, 2007 Inexact Methods for PDE-Constrained Optimization.
84 b Unidimensional Search Methods Most algorithms for unconstrained and constrained optimisation use an efficient unidimensional optimisation technique.
1 Jorge Nocedal Northwestern University With S. Hansen, R. Byrd and Y. Singer IPAM, UCLA, Feb 2014 A Stochastic Quasi-Newton Method for Large-Scale Learning.
1 Introduction to Linear and Nonlinear Programming.
General Nonlinear Programming (NLP) Software
Nonlinear programming Unconstrained optimization techniques.
Business Analytics with Nonlinear Programming
Optimization of Process Flowsheets S,S&L Chapter 24 T&S Chapter 12 Terry A. Ring CHEN 5253.
Linear Programming Chapter 6. Large Scale Optimization.
Survey of gradient based constrained optimization algorithms Select algorithms based on their popularity. Additional details and additional algorithms.
Frank Edward Curtis Northwestern University Joint work with Richard Byrd and Jorge Nocedal January 31, 2007 Inexact Methods for PDE-Constrained Optimization.
I N V E N T I V EI N V E N T I V E A Morphing Approach To Address Placement Stability Philip Chong Christian Szegedy.
Part 4 Nonlinear Programming 4.5 Quadratic Programming (QP)
A comparison between PROC NLP and PROC OPTMODEL Optimization Algorithm Chin Hwa Tan December 3, 2008.
Domain Decomposition in High-Level Parallelizaton of PDE codes Xing Cai University of Oslo.
Exact Differentiable Exterior Penalty for Linear Programming Olvi Mangasarian UW Madison & UCSD La Jolla Edward Wild UW Madison December 20, 2015 TexPoint.
Inexact SQP methods for equality constrained optimization Frank Edward Curtis Department of IE/MS, Northwestern University with Richard Byrd and Jorge.
Branch and Bound Algorithms Present by Tina Yang Qianmei Feng.
Optimization in Engineering Design 1 Introduction to Non-Linear Optimization.
Airline Optimization Problems Constraint Technologies International
OR Integer Programming ( 정수계획법 ). OR
Linear Programming Chapter 9. Interior Point Methods  Three major variants  Affine scaling algorithm - easy concept, good performance  Potential.
Recent Developments in Optimization and their Impact on Control Stephen Wright Argonne National Laboratory
Water resources planning and management by use of generalized Benders decomposition method to solve large-scale MINLP problems By Prof. André A. Keller.
Exact Differentiable Exterior Penalty for Linear Programming
Bounded Nonlinear Optimization to Fit a Model of Acoustic Foams
Xing Cai University of Oslo
Multiplicative updates for L1-regularized regression
Computational Optimization
Decomposed optimization-control problem
The New Faces of Nonlinear Programming
Jorge Nocedal Northwestern University
CS5321 Numerical Optimization
CS5321 Numerical Optimization
CS5321 Numerical Optimization
CS5321 Numerical Optimization
Part 4 Nonlinear Programming
Ph.D. Thesis Numerical Solution of PDEs and Their Object-oriented Parallel Implementations Xing Cai October 26, 1998.
Dr. Arslan Ornek DETERMINISTIC OPTIMIZATION MODELS
Part 4 Nonlinear Programming
Presentation transcript:

1 Algorithms and Software for Large-Scale Nonlinear Optimization OTC day, 6 Nov 2003 Richard Waltz, Northwestern University Project I: Large-scale Active-Set methods for NLP Fact or Fiction? (with J. Nocedal, R. Byrd and N. Gould) Project II: Adaptive Barrier Updates for NLP Interior- Point methods (with J. Nocedal, R. Byrd, and A. Waechter)

2 1. Successive Linear Programming (SLP) Inefficient, slow convergence 2. Successively Linearly Constrained (SLC) e.g. MINOS Difficulty scaling up 3. Sequential Quadratic Programming (SQP) e.g. filterSQP, SNOPT Very robust when less than a couple thousand degrees of freedom For larger problems QP subproblems may be too expensive Current Active-Set Methods

3 Fletcher, Sainz de la Maza (1989) Overview 0. Given: x 1. Solve LP to get working set W. 2. Compute a step, d, by solving an equality constrained QP using constraints in W. 3. Set: x T = x+d. SLP-EQP Approach

4 SLP-EQP Strengths: Only solve LP and EQP subproblems Early results very encouraging Competitive with SQP – able to solve problems with more degrees of freedom But… Not yet competitive with Interior Difficulties in warm starting LP subproblems How to handle degeneracy? Theory needs more development

5 NLP Functions twice continuously differentiable Adaptive barrier updates

6 Solve a sequence of barrier subproblems Approach solution to NLP as Adaptive barrier updates

7 Overview of Barrier Strategies: 1. Fixed decrease with barrier stop test (e.g. KNITRO) 2. Centrality-based strategies (e.g. LOQO) 3. Probing strategies (e.g. Mehrotra PC) Adaptive barrier updates (NLP)

8 KNITRO Conservative rule Initially  Decrease  linearly Fastlinear decrease near solution Globally convergent Robust but trade-off some efficiency Initial point option Adaptive barrier updates (NLP)

9 Develop a more flexible adaptive rule Allow increases in barrier parameter!   : function of: Spread of complementarity pairs Recent steplengths Ease of meeting a barrier stop test Probing step (e.g. predictor step) Adaptive barrier updates (NLP)

10 1. Official  for global conv (satisfies barrier stop test) 2. Trial  for flexibility    Globally Convergent Framework