Tutorials

Optimization of scalar functions

Standard Bayesian optimization

The target of the study is to run a standard Bayesian optimization of a scalar function.

Driver:

BayesianOptimization

Global minimization of a non-expensive scalar function

The target of the study is to run a global minimization a non-expensive scalar function without known gradient.

Driver:

DifferentialEvolution

Global minimization of a non-expensive scalar function with discrete parameters

The target of the study is to run a global minimization a non-expensive scalar function without known gradient on a mixed continuous and discrete design space.

Driver:

CMAES

Gradient-based minimization of a non-expensive scalar function

The target of the study is to run a minimization a scalar function subject to inequality constraints based on the scipy implementation of the SLSQP method.

Driver:

ScipyMinimizer

Least-squares optimization

Solution of least-square problem using Bayesian optimization

The target of the study is solve a non-linear least-squares problem based on the Bayesian optimization approach.

Driver:

BayesianLeastSquares

Solution of a non-expensive least-square problem based on a scipy implementation

The target of the study is solve a non-linear least-squares problem based on the scipy implementation of the Trust Region Reflective algorithm.

Driver:

ScipyLeastSquares

Bayesian parameter reconstruction using Bayesian optimization

The target of the study is solve the Bayesian parameter reconstrution problem based on the Bayesian optimization approach.

Driver:

BayesianReconstruction

Advanced Topics

Optimal control of a system in a changing environment

The target of the study is show how to control a system that depends on environment parameters such as temperature or humidity.

Driver:

ActiveLearning

Optimization of resonant system based on Gaussian fit

The target of the study is to tune a resonant system using a Gaussian fit of the amplitude as a function of the resonance wavelength. The study showcases the use of a fit variable.

Driver:

ActiveLearning

Benchmarking different studies against each other

Different minimization studies are benchmarked against each other for a specific objective function.

Drivers:

BayesianOptimization, ActiveLearning, CMAES, DifferentialEvolution, ScipyMinimizer

Variance-based sensitivity analysis for parameter reconstruction

A variance-based sensitivity analysis gives insights into how much information each measurement channel contributes for a parameter reconstruction.

Driver:

ActiveLearning

Multi-objective optimization

A multi-objective optimization allows to identify the trade-off between different objective values.

Driver:

ActiveLearning

Active learning of a global surrogate model

The target of the study is to train a surrogate model of a vectorial function in an active-learning loop.

Driver:

ActiveLearning