.. _method-multilevel_sampling-weighted: """""""" weighted """""""" Include control variate weights for each of the recursive differences using in multilevel Monte Carlo (MLMC) .. toctree:: :hidden: :maxdepth: 1 method-multilevel_sampling-weighted-solver_metric method-multilevel_sampling-weighted-sqp method-multilevel_sampling-weighted-nip method-multilevel_sampling-weighted-global_local method-multilevel_sampling-weighted-competed_local method-multilevel_sampling-weighted-search_model_graphs **Specification** - *Alias:* None - *Arguments:* None - *JSON argument key:* None **Child Keywords:** +-------------------------+------------------------+---------------------+--------------------------------+------------------------------------------------------------+ | Required/Optional | Description of Group | JSON Group Key | Dakota Keyword | Dakota Keyword Description | +=========================+========================+=====================+================================+============================================================+ | Optional | | | `solver_metric`__ | Metric employed during numerical solutions in | | | | | | sampling-based multifidelity UQ methods. | +-------------------------+------------------------+---------------------+--------------------------------+------------------------------------------------------------+ | Optional (Choose One) | Optimization Solver | optimization_solver | `sqp`__ | Use a sequential quadratic programming method for solving | | | | | | an optimization sub-problem | + + + +--------------------------------+------------------------------------------------------------+ | | | | `nip`__ | Use a nonlinear interior point method for solving an | | | | | | optimization sub-problem | + + + +--------------------------------+------------------------------------------------------------+ | | | | `global_local`__ | Use a hybrid global-local scheme for solving an | | | | | | optimization sub-problem | + + + +--------------------------------+------------------------------------------------------------+ | | | | `competed_local`__ | Use a competed local solver scheme for solving an | | | | | | optimization sub-problem | +-------------------------+------------------------+---------------------+--------------------------------+------------------------------------------------------------+ | Optional | | | `search_model_graphs`__ | For weighted multilevel Monte Carlo, this option activates | | | | | | a search over possible hierarchical model graphs | +-------------------------+------------------------+---------------------+--------------------------------+------------------------------------------------------------+ .. __: method-multilevel_sampling-weighted-solver_metric.html __ method-multilevel_sampling-weighted-sqp.html __ method-multilevel_sampling-weighted-nip.html __ method-multilevel_sampling-weighted-global_local.html __ method-multilevel_sampling-weighted-competed_local.html __ method-multilevel_sampling-weighted-search_model_graphs.html **Description** Referring to generalized ACV (:dakkw:`method-approximate_control_variate-search_model_graphs` and :cite:p:`Bomarito2022`), weighted MLMC is a special case of generalized ACV-RD (:dakkw:`method-approximate_control_variate-acv_recursive_diff`) where a hierarchical DAG is employed across the model approximations. As such, a weighted MLMC specification forwards to the generalized ACV solver, but with fixing the DAG to be hierarchical (each approximation node points to the next approximation of higher fidelity/resolution, ending with the truth model at the root node) and fixing the sampling scheme to be ACV-RD. While the use of a hierarchical DAG is required, the approximation selections and orderings within this DAG can be varied, so generalized ACV capabilities for model graph search (different hierarchical orderings) and model selection (different approximation subsets) are available for a specification of weighted MLMC -- see :dakkw:`method-multilevel_sampling-weighted-search_model_graphs`. **Theory** Refer to :cite:p:`Bomarito2022` for understanding ACV generalizations for the different control variate pairings that are possible when codified into a more comprehensive set of potential DAGs.