Optimization via Random Search — mlr_optimizers_random

OptimizerBatchRandomSearch class that implements a simple Random Search.

In order to support general termination criteria and parallelization, we evaluate points in a batch-fashion of size batch_size. Larger batches mean we can parallelize more, smaller batches imply a more fine-grained checking of termination criteria.

Source

Bergstra J, Bengio Y (2012). “Random Search for Hyper-Parameter Optimization.” Journal of Machine Learning Research, 13(10), 281–305. https://jmlr.csail.mit.edu/papers/v13/bergstra12a.html.

Dictionary

This Optimizer can be instantiated via the dictionary mlr_optimizers or with the associated sugar function opt():

mlr_optimizers$get("random_search")
opt("random_search")

Parameters

batch_size: integer(1)
Maximum number of points to try in a batch.

Progress Bars

$optimize() supports progress bars via the package progressr combined with a Terminator. Simply wrap the function in progressr::with_progress() to enable them. We recommend to use package progress as backend; enable with progressr::handlers("progress").

Super classes

Optimizer -> OptimizerBatch -> OptimizerBatchRandomSearch

Methods

Inherited methods

`OptimizerBatchRandomSearch$new()`

Creates a new instance of this R6 class.

Usage

OptimizerBatchRandomSearch$new()

`OptimizerBatchRandomSearch$clone()`

The objects of this class are cloneable with this method.

Usage

OptimizerBatchRandomSearch$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Examples

# define the objective function
fun = function(xs) {
  list(y = - (xs[[1]] - 2)^2 - (xs[[2]] + 3)^2 + 10)
}

# set domain
domain = ps(
  x1 = p_dbl(-10, 10),
  x2 = p_dbl(-5, 5)
)

# set codomain
codomain = ps(
  y = p_dbl(tags = "maximize")
)

# create objective
objective = ObjectiveRFun$new(
  fun = fun,
  domain = domain,
  codomain = codomain,
  properties = "deterministic"
)

# initialize instance
instance = oi(
  objective = objective,
  terminator = trm("evals", n_evals = 20)
)

# load optimizer
optimizer = opt("random_search", batch_size = 10)

# trigger optimization
optimizer$optimize(instance)
#>          x1        x2  x_domain        y
#>       <num>     <num>    <list>    <num>
#> 1: 1.388184 -4.170693 <list[2]> 8.255159

# all evaluated configurations
instance$archive
#> 
#> ── <ArchiveBatch> - Data Table Storage ─────────────────────────────────────────
#>        x1    x2     y           timestamp batch_nr x_domain_x1 x_domain_x2
#>     <num> <num> <num>              <POSc>    <int>       <num>       <num>
#>  1: -0.40   0.8   -10 2026-09-18 09:12:30        1       -0.40         0.8
#>  2: -4.16  -3.4   -28 2026-09-18 09:12:30        1       -4.16        -3.4
#>  3:  1.85  -4.7     7 2026-09-18 09:12:30        1        1.85        -4.7
#>  4:  8.04  -1.5   -29 2026-09-18 09:12:30        1        8.04        -1.5
#>  5: -0.62   1.9   -21 2026-09-18 09:12:30        1       -0.62         1.9
#>  6: -5.04  -1.3   -42 2026-09-18 09:12:30        1       -5.04        -1.3
#>  7: -6.52   2.4   -92 2026-09-18 09:12:30        1       -6.52         2.4
#>  8:  8.87   0.8   -51 2026-09-18 09:12:30        1        8.87         0.8
#>  9:  0.56  -2.0     7 2026-09-18 09:12:30        1        0.56        -2.0
#> 10:  5.98  -3.5    -6 2026-09-18 09:12:30        1        5.98        -3.5
#> 11:  7.95   3.8   -72 2026-09-18 09:12:30        2        7.95         3.8
#> 12: -0.07  -4.0     5 2026-09-18 09:12:30        2       -0.07        -4.0
#> 13:  1.39  -4.2     8 2026-09-18 09:12:30        2        1.39        -4.2
#> 14:  9.31   0.8   -58 2026-09-18 09:12:30        2        9.31         0.8
#> 15:  0.12  -1.6     5 2026-09-18 09:12:30        2        0.12        -1.6
#> 16:  6.01   1.5   -27 2026-09-18 09:12:30        2        6.01         1.5
#> 17: -2.40   3.7   -54 2026-09-18 09:12:30        2       -2.40         3.7
#> 18:  1.61   3.9   -38 2026-09-18 09:12:30        2        1.61         3.9
#> 19:  0.49   0.6    -5 2026-09-18 09:12:30        2        0.49         0.6
#> 20: -5.54   2.1   -73 2026-09-18 09:12:30        2       -5.54         2.1
#>        x1    x2     y           timestamp batch_nr x_domain_x1 x_domain_x2
#>     <num> <num> <num>              <POSc>    <int>       <num>       <num>

# best performing configuration
instance$result
#>          x1        x2  x_domain        y
#>       <num>     <num>    <list>    <num>
#> 1: 1.388184 -4.170693 <list[2]> 8.255159