Run Active Learning — optimize

Convenience function that constructs an active learning mlr3mbo::OptimizerMbo via optimizer_active_learning(), runs it on a bbotk instance, and (optionally) logs metrics via CallbackMetricsTracker.

Usage

optimize_active(
  objective,
  search_space = NULL,
  term_evals = NULL,
  terminator = NULL,
  metrics_tracker = NULL,
  ...
)

objective: (bbotk::Objective)
Objective to evaluate. Typically has a single codomain target tagged "learn".
search_space: (NULL | paradox::ParamSet)
Optional restricted search space. If NULL, the search space is derived from objective$domain (same logic as bbotk's OptimInstanceBatch).
term_evals: (NULL | integer(1))
Convenience evaluation budget used only if terminator is NULL.
terminator: (NULL | bbotk::Terminator)
Terminator for the outer active learning loop. If NULL, a trm("evals", n_evals = term_evals) is constructed.
metrics_tracker: (NULL | MetricsTracker)
Optional metrics tracker. If provided, a CallbackMetricsTracker is attached to the instance.
...: Passed to optimizer_active_learning().

list() with: