.. _model-surrogate-multipoint: """""""""" multipoint """""""""" Construct a surrogate from multiple existing training points .. toctree:: :hidden: :maxdepth: 1 model-surrogate-multipoint-tana model-surrogate-multipoint-qmea model-surrogate-multipoint-truth_model_pointer **Specification** - *Alias:* None - *Arguments:* None **Child Keywords:** +-------------------------+--------------------+-------------------------+---------------------------------------------+ | Required/Optional | Description of | Dakota Keyword | Dakota Keyword Description | | | Group | | | +=========================+====================+=========================+=============================================+ | Required (Choose One) | Multipoint | `tana`__ | Local multi-point model via two-point | | | Surrogate | | nonlinear approximation | | | +-------------------------+---------------------------------------------+ | | | `qmea`__ | Multi-point surrogate approximation based | | | | | on QMEA algorithm | +-------------------------+--------------------+-------------------------+---------------------------------------------+ | Required | `truth_model_pointer`__ | Pointer to specify a "truth" model, from | | | | which to construct a surrogate | +----------------------------------------------+-------------------------+---------------------------------------------+ .. __: model-surrogate-multipoint-tana.html __ model-surrogate-multipoint-qmea.html __ model-surrogate-multipoint-truth_model_pointer.html **Description** Multipoint approximations use data from previous design points to improve the accuracy of local approximations. The data often comes from the current and previous iterates of a minimization algorithm. Currently, only the Two-point Adaptive Nonlinearity Approximation (TANA-3) method of :cite:p:`Xu98` is supported with the ``tana`` keyword. The truth model to be used to generate the value/gradient data used in the approximation is identified through the required ``truth_model_pointer`` specification.