nowpac

Gradient-free inequality-constrained optimization using Nonlinear Optimization With Path Augmented Constraints (NOWPAC).

Specification

  • Alias: None

  • Arguments: None

Child Keywords:

Required/Optional

Description of Group

Dakota Keyword

Dakota Keyword Description

Optional

trust_region

Use trust region as the globalization strategy.

Optional

max_iterations

Number of iterations allowed for optimizers and adaptive UQ methods

Optional

max_function_evaluations

Number of function evaluations allowed for optimizers

Optional

scaling

Turn on scaling for variables, responses, and constraints

Optional

model_pointer

Identifier for model block to be used by a method

Description

NOWPAC is a provably-convergent gradient-free optimization method from MIT that solves a series of trust region surrogate-based subproblems to generate improving steps. Due to its use of an interior penalty scheme and enforcement of strict feasibility, it does not support linear or nonlinear equality constraints. As opposed to the stochastic version (SNOWPAC), NOWPAC does not currently support a feasibility restoration mode, so it is necessary to start from a feasible design.

Note: (S)NOWPAC is not configured with Dakota by default and requires a separate installation of the NOWPAC distribution from MIT, combined with its TPLs of Eigen and NLOPT.

Examples

method
  nowpac
    max_function_evaluations = 1000
    convergence_tolerance = 1e-4
    trust_region
      initial_size = 0.10
      minimum_size = 1.0e-6
      contract_threshold = 0.25
      expand_threshold   = 0.75
      contraction_factor = 0.50
      expansion_factor   = 1.50