Optimization via Random Search

OptimizerRandomSearch 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 class

bbotk::Optimizer -> OptimizerRandomSearch

Methods

Public methods

• OptimizerRandomSearch$new()

• OptimizerRandomSearch$clone()

Inherited methods

• bbotk::Optimizer$format()
• bbotk::Optimizer$help()
• bbotk::Optimizer$optimize()
• bbotk::Optimizer$print()

---

Method new()

Creates a new instance of this R6 class.

Usage

OptimizerRandomSearch$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

OptimizerRandomSearch$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

search_space = domain = ps(x = p_dbl(lower = -1, upper = 1))

codomain = ps(y = p_dbl(tags = "minimize"))

objective_function = function(xs) {
  list(y = as.numeric(xs)^2)
}

objective = ObjectiveRFun$new(
 fun = objective_function,
 domain = domain,
 codomain = codomain)

instance = OptimInstanceSingleCrit$new(
 objective = objective,
 search_space = search_space,
 terminator = trm("evals", n_evals = 10))


optimizer = opt("random_search")

# modifies the instance by reference
optimizer$optimize(instance)
#>              x  x_domain            y
#>          <num>    <list>        <num>
#> 1: -0.01882196 <list[1]> 0.0003542662

# returns best scoring evaluation
instance$result
#>              x  x_domain            y
#>          <num>    <list>        <num>
#> 1: -0.01882196 <list[1]> 0.0003542662

# allows access of data.table of full path of all evaluations
as.data.table(instance$archive$data)
#>               x            y  x_domain           timestamp batch_nr
#>           <num>        <num>    <list>              <POSc>    <int>
#>  1: -0.28203532 0.0795439212 <list[1]> 2024-02-29 15:30:07        1
#>  2:  0.36047183 0.1299399404 <list[1]> 2024-02-29 15:30:07        2
#>  3: -0.57370263 0.3291347057 <list[1]> 2024-02-29 15:30:07        3
#>  4: -0.01882196 0.0003542662 <list[1]> 2024-02-29 15:30:07        4
#>  5:  0.92688144 0.8591092090 <list[1]> 2024-02-29 15:30:07        5
#>  6:  0.89315148 0.7977195600 <list[1]> 2024-02-29 15:30:07        6
#>  7: -0.89623084 0.8032297230 <list[1]> 2024-02-29 15:30:07        7
#>  8: -0.40985141 0.1679781748 <list[1]> 2024-02-29 15:30:07        8
#>  9:  0.24838006 0.0616926532 <list[1]> 2024-02-29 15:30:07        9
#> 10:  0.14358194 0.0206157735 <list[1]> 2024-02-29 15:30:07       10