Optimization of Algorithmic Parameters using a Meta-Control Approach*

Authors:
Wolf Kohn;Zelda B. Zabinsky;Vladimir Brayman
Affiliations:
Clearsight Systems Inc., Industrial Engineering, University of Washington, Bellevue, USA 98006;Clearsight Systems Inc., Industrial Engineering, University of Washington, Bellevue, USA 98006;Clearsight Systems Inc., Industrial Engineering, University of Washington, Bellevue, USA 98006
Venue:
Journal of Global Optimization
Year:
2006

Citing 10
Cited 2

Nonlinear programming: theory, algorithms, and applications

Nonlinear programming: theory, algorithms, and applications
More test examples for nonlinear programming codes

More test examples for nonlinear programming codes
Interior path following primal-dual algorithms. Part I: Linear programming

Mathematical Programming: Series A and B
Interior path following primal-dual algorithms. Part II: Convex quadratic programming

Mathematical Programming: Series A and B
Hybrid systems: chattering approximation to relaxed controls

Proceedings of the DIMACS/SYCON workshop on Hybrid systems III : verification and control: verification and control
Numerical methods for ordinary differential equations in the 20th century

Journal of Computational and Applied Mathematics - Special issue on numerical anaylsis 2000 Vol. VI: Ordinary differential equations and integral equations
Computational Differential Equations

Computational Differential Equations
Linear Optimal Control Systems

Linear Optimal Control Systems
Interior Methods for Nonlinear Optimization

SIAM Review
Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)

Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)

The interacting-particle algorithm with dynamic heating and cooling

Journal of Global Optimization
On sample size control in sample average approximations for solving smooth stochastic programs

Computational Optimization and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Optimization algorithms usually rely on the setting of parameters, such as barrier coefficients. We have developed a generic meta-control procedure to optimize the behavior of given iterative optimization algorithms. In this procedure, an optimal continuous control problem is defined to compute the parameters of an iterative algorithm as control variables to achieve a desired behavior of the algorithm (e.g., convergence time, memory resources, and quality of solution). The procedure is illustrated with an interior point algorithm to control barrier coefficients for constrained nonlinear optimization. Three numerical examples are included to demonstrate the enhanced performance of this method.