Dakota Logo

Install

  • Download and Install Dakota

Getting Started with Dakota

  • What is Dakota?
  • Anatomy of a Dakota Study
  • Tutorial: Your First Dakota Study
  • Connecting Your Code: The Analysis Driver

Using Dakota

  • Using Dakota
    • Examples
    • Dakota Input File
    • Library APIs
    • Running Dakota
    • Dakota Output
    • Study Types
    • Topics
    • Advanced Topics
    • Dakota Theory
    • Keyword Reference
      • environment
      • method
        • adaptive_sampling
        • approximate_control_variate
        • asynch_pattern_search
        • bayes_calibration
        • branch_and_bound
        • centered_parameter_study
        • coliny_beta
        • coliny_cobyla
        • coliny_direct
        • coliny_ea
        • coliny_pattern_search
        • coliny_solis_wets
        • conmin_frcg
        • conmin_mfd
        • dace
        • demo_tpl
        • dot_bfgs
        • dot_frcg
        • dot_mmfd
        • dot_slp
        • dot_sqp
        • efficient_global
        • external_python
        • fsu_cvt
        • fsu_quasi_mc
        • function_train
        • genie_direct
        • genie_opt_darts
        • global_evidence
        • global_interval_est
        • global_reliability
        • gpais
        • hybrid_sequential
        • import_points
        • importance_sampling
        • list_parameter_study
        • local_evidence
        • local_interval_est
        • local_reliability
        • mesh_adaptive_search
          • model_pointer
          • scaling
          • max_function_evaluations
          • max_iterations
          • id_method
          • output
          • final_solutions
          • initial_delta
          • variable_tolerance
          • function_precision
          • seed
          • history_file
          • display_format
          • variable_neighborhood_search
          • neighbor_order
          • display_all_evaluations
          • use_surrogate
        • moga
        • multi_start
        • multidim_parameter_study
        • multifidelity_function_train
        • multifidelity_polynomial_chaos
        • multifidelity_sampling
        • multifidelity_stoch_collocation
        • multilevel_blue
        • multilevel_function_train
        • multilevel_multifidelity_sampling
        • multilevel_polynomial_chaos
        • multilevel_sampling
        • ncsu_direct
        • nl2sol
        • nlpql_sqp
        • nlssol_sqp
        • nonlinear_cg
        • nowpac
        • npsol_sqp
        • optpp_cg
        • optpp_fd_newton
        • optpp_g_newton
        • optpp_newton
        • optpp_pds
        • optpp_q_newton
        • pareto_set
        • pof_darts
        • polynomial_chaos
        • psuade_moat
        • richardson_extrap
        • rkd_darts
        • rol
        • sampling
        • snowpac
        • soga
        • stoch_collocation
        • surrogate_based_global
        • surrogate_based_local
        • surrogate_based_uq
        • vector_parameter_study
      • model
      • variables
      • interface
      • responses
      • Keyword Pages
  • Using Dakota GUI
  • Using External Tools

Building and Developing Dakota

  • Building Dakota
  • Developing Dakota

More

  • Miscellaneous
Dakota

Sandia National Laboratories

  • Search Sandia.gov
  • All Sandia Websites
  • Using Dakota
  • Keyword Reference
  • method
  • mesh_adaptive_search
  • variable_neighborhood_search
  • View page source

variable_neighborhood_search

Percentage of evaluations to do to escape local minima.

Specification

  • Alias: None

  • Arguments: None

  • JSON argument key: None

  • Default: 0.0

Description

The variable_neighborhood_search keyword is used to set the percentage (in decimal form) of function evaluations used to escape local minima. The mesh adaptive direct search method will try to perform a maximum of that percentage of the function evaluations within this more extensive search.

Default Behavior

By default, variable_neighborhood_search is not used.

Usage Tips

Using variable_neighborhood_search results in an increased number of function evaluations. If the desired result is a local minimum, the added cost is of little or no value, so the recommendation is not to use it. If the desired result is the best local minimum possible within a computational budget, then there is value in setting this parameter. Note that the higher the value, the greater the computational cost. The value should be no greater than 1.0.

Examples

The following example shows the syntax used to set variable_neighborhood_search.

method
  mesh_adaptive_search
    seed = 1234
    variable_neighborhood_search = 0.1
{
  "method": [
    {
      "mesh_adaptive_search": {
        "seed": 1234,
        "variable_neighborhood_search": 0.1
      }
    }
  ]
}
Previous Next

Exceptional service in the national interest

© 2023 National Technology and Engineering Solutions of Sandia, LLC. | Questions & Comments | Privacy & Security

U.S. Department of Energy National Nuclear Security Administration

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA-0003525.