Surrogate Optimization

Surrogate optimization solver for expensive objective functions, with bounds and optional integer constraints

Use surrogate optimization for expensive (time-consuming) objective functions. The solver requires finite bounds on all variables, allows for nonlinear inequality constraints, and accepts integer constraints on selected variables. The solver can optionally save its state after each function evaluation, enabling recovery from premature stops.

Functions

expand all

Problem-Based Solution

`optimvalues`	Create values for optimization problem
`solve`	Solve optimization problem or equation problem

Solver

surrogateopt Surrogate optimization for global minimization of time-consuming objective functions

Convert Constraints

packfcn Combine objective and nonlinear constraint functions

Options

`optimoptions`	Create optimization options
`resetoptions`	Reset options

Live Editor Tasks

Optimize

Optimize or solve equations in the Live Editor

Topics

Problem-Based Surrogate Optimization

Optimize Multidimensional Function Using surrogateopt, Problem-Based
Basic example minimizing a multidimensional function in the problem-based approach.
Mixed-Integer Surrogate Optimization, Problem-Based
Solve integer and mixed-integer problems using the problem-based approach and surrogateopt.
Specify Starting Points and Values for surrogateopt, Problem-Based
Specify start points and their function values using optimvalues in the problem-based approach.
Solve Feasibility Problem Using surrogateopt, Problem-Based
Solve a feasibility problem using the problem-based approach and surrogateopt solver.
Feasibility Using Problem-Based Optimize Live Editor Task
Solve a nonlinear feasibility problem using the problem-based Optimize Live Editor task and several solvers.

Optimize Using Surrogate Optimization

Surrogate Optimization of Multidimensional Function
Solve a multidimensional problem using surrogateopt, patternsearch, and fmincon, and then compare the results.
Modify surrogateopt Options
Search for the global minimum using surrogateopt, and then modify options of the function to revise the search.
Interpret surrogateoptplot
How to interpret a surrogateoptplot plot.
Compare Surrogate Optimization with Other Solvers
Compare surrogateopt to patternsearch and fmincon on a nonsmooth problem.
Surrogate Optimization of Six-Element Yagi-Uda Antenna
Solve an antenna design problem using surrogate optimization.
Work with Checkpoint Files
Shows how to use checkpoint files to restart, recover, analyze, or extend an optimization.
Surrogate Optimization with Nonlinear Constraint
Solve a problem containing a nonlinear ODE with a nonlinear constraint using surrogateopt.
Convert Nonlinear Constraints Between surrogateopt Form and Other Solver Forms
Presents techniques for converting objective and nonlinear constraint functions for other solvers to and from surrogateopt form.
Mixed-Integer Surrogate Optimization
Integer-constrained surrogate optimization.
Optimal Component Choice Using surrogateopt
Choose components from lists to best fit a response curve.
Solve Nonlinear Problem with Integer and Nonlinear Constraints
Compare the solution of a nonlinear problem both with and without integer constraints.
Solve Feasibility Problem
Use surrogateopt to solve a feasibility problem.
Fix Variables in surrogateopt
Fix some variables by setting their upper and lower bounds equal.
Vectorized Surrogate Optimization for Custom Parallel Simulation
This example shows how to perform custom parallel optimization using the surrogateopt UseVectorized and BatchUpdateInterval options.
Improve surrogateopt Solution or Process
Hints for obtaining a better solution or obtaining a solution more quickly.

Surrogate Optimization Background

What Is Surrogate Optimization?
Surrogate optimization attempts to find a global minimum of an objective function using few objective function evaluations.
Surrogate Optimization Algorithm
Learn details of the surrogate optimization algorithm, when run in serial or parallel.
Surrogate Optimization Options
Explore the options for surrogate optimization, including algorithm control, stopping criteria, command-line display, and output and plot functions.

Related Information