mlrMBO v1.1.2


Monthly downloads



Bayesian Optimization and Model-Based Optimization of Expensive Black-Box Functions

Flexible and comprehensive R toolbox for model-based optimization ('MBO'), also known as Bayesian optimization. It implements the Efficient Global Optimization Algorithm and is designed for both single- and multi- objective optimization with mixed continuous, categorical and conditional parameters. The machine learning toolbox 'mlr' provide dozens of regression learners to model the performance of the target algorithm with respect to the parameter settings. It provides many different infill criteria to guide the search process. Additional features include multi-point batch proposal, parallel execution as well as visualization and sophisticated logging mechanisms, which is especially useful for teaching and understanding of algorithm behavior. 'mlrMBO' is implemented in a modular fashion, such that single components can be easily replaced or adapted by the user for specific use cases.



CRAN_Status_Badge Build Status Build status Coverage Status Monthly RStudio CRAN Downloads

Model-based optimization with mlr.


We reccomend to install the official release version:


For experimental use you can install the latest development version:



MBO demo

mlrMBO is a highly configurable R toolbox for model-based / Bayesian optimization of black-box functions.


  • EGO-type algorithms (Kriging with expected improvement) on purely numerical search spaces, see Jones et al. (1998)
  • Mixed search spaces with numerical, integer, categorical and subordinate parameters
  • Arbitrary parameter transformation allowing to optimize on, e.g., logscale
  • Optimization of noisy objective functions
  • Multi-Criteria optimization with approximated Pareto fronts
  • Parallelization through multi-point batch proposals
  • Parallelization on many parallel back-ends and clusters through batchtools and parallelMap

For the surrogate, mlrMBO allows any regression learner from mlr, including:

  • Kriging aka. Gaussian processes (i.e. DiceKriging)
  • random Forests (i.e. randomForest)
  • and many more...

Various infill criteria (aka. acquisition functions) are available:

  • Expected improvement (EI)
  • Upper/Lower confidence bound (LCB, aka. statistical lower or upper bound)
  • Augmented expected improvement (AEI)
  • Expected quantile improvement (EQI)
  • API for custom infill criteria

Objective functions are created with package smoof, which also offers many test functions for example runs or benchmarks.

Parameter spaces and initial designs are created with package ParamHelpers.

mlrMBO - How to Cite and Citing Publications

Please cite our arxiv paper (Preprint). You can get citation info via citation("mlrMBO") or copy the following BibTex entry:

  title = {{{mlrMBO}}: {{A Modular Framework}} for {{Model}}-{{Based Optimization}} of {{Expensive Black}}-{{Box Functions}}},
  url = {},
  shorttitle = {{{mlrMBO}}},
  archivePrefix = {arXiv},
  eprinttype = {arxiv},
  eprint = {1703.03373},
  primaryClass = {stat},
  author = {Bischl, Bernd and Richter, Jakob and Bossek, Jakob and Horn, Daniel and Thomas, Janek and Lang, Michel},
  date = {2017-03-09},

Some parts of the package were created as part of other publications. If you use these parts, please cite the relevant work appropriately:

Functions in mlrMBO

Name Description
error_handling Error handling for mlrMBO
makeMBOInfillCrit Create an infill criterion.
MBOMultiObjResult Multi-Objective result object.
proposePoints Propose candidates for the objective function
getSupportedInfillOptFunctions Get names of supported infill-criteria optimizers.
getSupportedMultipointInfillOptFunctions Get names of supported multi-point infill-criteria optimizers.
plotMBOResult MBO Result Plotting
setMBOControlInfill Extends mbo control object with infill criteria and infill optimizer options.
setMBOControlMultiObj Set multi-objective options.
renderExampleRunPlot Renders plots for exampleRun objects, either in 1D or 2D, or exampleRunMultiObj objects.
print.MBOControl Print mbo control object.
exampleRunMultiObj Perform an MBO run on a multi-objective test function and and visualize what happens.
finalizeSMBO Finalizes the SMBO Optimization
makeMBOLearner Generate default learner.
OptResult OptResult object.
makeMBOTrafoFunction Create a transformation function for MBOExampleRun.
OptState OptState object.
getGlobalOpt Helper function which returns the (estimated) global optimum.
plot.OptState Generate ggplot2 Object
getMBOInfillCrit Get properties of MBO infill criterion.
MBOSingleObjResult Single-Objective result object.
mbo Optimizes a function with sequential model based optimization.
mboContinue Continues an mbo run from a save-file.
plotExampleRun Renders plots for exampleRun objects and displays them.
OptProblem OptProblem object.
initSMBO Initialize a manual sequential MBO run.
makeMBOControl Set MBO options.
mbo_parallel Parallelization in mlrMBO
trafos Transformation methods.
setMBOControlMultiPoint Set multipoint proposal options.
updateSMBO Updates SMBO with the new observations
mlrMBO_examples mlrMBO examples
setMBOControlTermination Set termination options.
exampleRun Perform an mbo run on a test function and and visualize what happens.
infillcrits Infill criteria.
initCrit Initialize an MBO infill criterion.
mboFinalize Finalizes an mbo run from a save-file.
mbo_OptPath OptPath in mlrMBO
No Results!

Vignettes of mlrMBO

No Results!

Last month downloads


License BSD_2_clause + file LICENSE
LazyData yes
Encoding UTF-8
ByteCompile yes
RoxygenNote 6.0.1
VignetteBuilder knitr
NeedsCompilation yes
Packaged 2018-06-21 07:41:18 UTC; richter
Repository CRAN
Date/Publication 2018-06-21 16:52:29 UTC

Include our badge in your README