Dakota  Version 6.4
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Pages
Protected Member Functions | Protected Attributes | List of all members
SurrBasedMinimizer Class Reference

Base class for local/global surrogate-based optimization/least squares. More...

Inheritance diagram for SurrBasedMinimizer:
Minimizer Iterator EffGlobalMinimizer SurrBasedGlobalMinimizer SurrBasedLocalMinimizer

Protected Member Functions

 SurrBasedMinimizer (ProblemDescDB &problem_db, Model &model)
 constructor
 
 ~SurrBasedMinimizer ()
 destructor
 
void derived_init_communicators (ParLevLIter pl_iter)
 derived class contributions to initializing the communicators associated with this Iterator instance
 
void derived_set_communicators (ParLevLIter pl_iter)
 derived class contributions to setting the communicators associated with this Iterator instance
 
void derived_free_communicators (ParLevLIter pl_iter)
 derived class contributions to freeing the communicators associated with this Iterator instance
 
void initialize_graphics (int iterator_server_id=1)
 initialize graphics customized for surrogate-based iteration
 
void print_results (std::ostream &s)
 
void update_lagrange_multipliers (const RealVector &fn_vals, const RealMatrix &fn_grads)
 initialize and update Lagrange multipliers for basic Lagrangian More...
 
void update_augmented_lagrange_multipliers (const RealVector &fn_vals)
 initialize and update the Lagrange multipliers for augmented Lagrangian More...
 
bool update_filter (const RealVector &fn_vals)
 update a filter from a set of function values More...
 
Real lagrangian_merit (const RealVector &fn_vals, const BoolDeque &sense, const RealVector &primary_wts, const RealVector &nln_ineq_l_bnds, const RealVector &nln_ineq_u_bnds, const RealVector &nln_eq_tgts)
 compute a Lagrangian function from a set of function values More...
 
void lagrangian_gradient (const RealVector &fn_vals, const RealMatrix &fn_grads, const BoolDeque &sense, const RealVector &primary_wts, const RealVector &nln_ineq_l_bnds, const RealVector &nln_ineq_u_bnds, const RealVector &nln_eq_tgts, RealVector &lag_grad)
 compute the gradient of the Lagrangian function
 
Real augmented_lagrangian_merit (const RealVector &fn_vals, const BoolDeque &sense, const RealVector &primary_wts, const RealVector &nln_ineq_l_bnds, const RealVector &nln_ineq_u_bnds, const RealVector &nln_eq_tgts)
 compute an augmented Lagrangian function from a set of function values More...
 
void augmented_lagrangian_gradient (const RealVector &fn_vals, const RealMatrix &fn_grads, const BoolDeque &sense, const RealVector &primary_wts, const RealVector &nln_ineq_l_bnds, const RealVector &nln_ineq_u_bnds, const RealVector &nln_eq_tgts, RealVector &alag_grad)
 compute the gradient of the augmented Lagrangian function
 
Real penalty_merit (const RealVector &fn_vals, const BoolDeque &sense, const RealVector &primary_wts)
 compute a penalty function from a set of function values More...
 
void penalty_gradient (const RealVector &fn_vals, const RealMatrix &fn_grads, const BoolDeque &sense, const RealVector &primary_wts, RealVector &pen_grad)
 compute the gradient of the penalty function
 
Real constraint_violation (const RealVector &fn_vals, const Real &constraint_tol)
 compute the constraint violation from a set of function values More...
 
- Protected Member Functions inherited from Minimizer
 Minimizer ()
 default constructor
 
 Minimizer (ProblemDescDB &problem_db, Model &model)
 standard constructor More...
 
 Minimizer (unsigned short method_name, Model &model)
 alternate constructor for "on the fly" instantiations
 
 Minimizer (unsigned short method_name, size_t num_lin_ineq, size_t num_lin_eq, size_t num_nln_ineq, size_t num_nln_eq)
 alternate constructor for "on the fly" instantiations
 
 ~Minimizer ()
 destructor
 
void update_from_model (const Model &model)
 set inherited data attributes based on extractions from incoming model
 
void initialize_run ()
 utility function to perform common operations prior to pre_run(); typically memory initialization; setting of instance pointers More...
 
void post_run (std::ostream &s)
 post-run portion of run (optional); verbose to print results; re-implemented by Iterators that can read all Variables/Responses and perform final analysis phase in a standalone way More...
 
void finalize_run ()
 utility function to perform common operations following post_run(); deallocation and resetting of instance pointers More...
 
const Modelalgorithm_space_model () const
 
Model original_model (unsigned short recasts_left=0)
 Return a shallow copy of the original model this Iterator was originally passed, optionally leaving recasts_left on top of it.
 
void data_transform_model ()
 Wrap iteratedModel in a RecastModel that subtracts provided observed data from the primary response functions (variables and secondary responses are unchanged) More...
 
void scale_model ()
 Wrap iteratedModel in a RecastModel that performs variable and/or response scaling. More...
 
Real objective (const RealVector &fn_vals, const BoolDeque &max_sense, const RealVector &primary_wts) const
 compute a composite objective value from one or more primary functions More...
 
Real objective (const RealVector &fn_vals, size_t num_fns, const BoolDeque &max_sense, const RealVector &primary_wts) const
 compute a composite objective with specified number of source primary functions, instead of userPrimaryFns More...
 
void objective_gradient (const RealVector &fn_vals, const RealMatrix &fn_grads, const BoolDeque &max_sense, const RealVector &primary_wts, RealVector &obj_grad) const
 compute the gradient of the composite objective function
 
void objective_gradient (const RealVector &fn_vals, size_t num_fns, const RealMatrix &fn_grads, const BoolDeque &max_sense, const RealVector &primary_wts, RealVector &obj_grad) const
 compute the gradient of the composite objective function More...
 
void objective_hessian (const RealVector &fn_vals, const RealMatrix &fn_grads, const RealSymMatrixArray &fn_hessians, const BoolDeque &max_sense, const RealVector &primary_wts, RealSymMatrix &obj_hess) const
 compute the Hessian of the composite objective function
 
void objective_hessian (const RealVector &fn_vals, size_t num_fns, const RealMatrix &fn_grads, const RealSymMatrixArray &fn_hessians, const BoolDeque &max_sense, const RealVector &primary_wts, RealSymMatrix &obj_hess) const
 compute the Hessian of the composite objective function More...
 
void archive_allocate_best (size_t num_points)
 allocate results arrays and labels for multipoint storage
 
void archive_best (size_t index, const Variables &best_vars, const Response &best_resp)
 archive the best point into the results array
 
void resize_best_vars_array (size_t newsize)
 Safely resize the best variables array to newsize taking into account the envelope-letter design pattern and any recasting. More...
 
void resize_best_resp_array (size_t newsize)
 Safely resize the best response array to newsize taking into account the envelope-letter design pattern and any recasting. More...
 
Real sum_squared_residuals (size_t num_pri_fns, const RealVector &residuals, const RealVector &weights)
 return weighted sum of squared residuals
 
void print_residuals (size_t num_terms, const RealVector &best_terms, const RealVector &weights, size_t num_best, size_t best_index, std::ostream &s)
 print num_terms residuals and misfit for final results
 
void print_model_resp (size_t num_pri_fns, const RealVector &best_fns, size_t num_best, size_t best_index, std::ostream &s)
 print the original user model resp in the case of data transformations
 
void local_recast_retrieve (const Variables &vars, Response &response) const
 infers MOO/NLS solution from the solution of a single-objective optimizer More...
 
- Protected Member Functions inherited from Iterator
 Iterator (BaseConstructor, ProblemDescDB &problem_db)
 constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139) More...
 
 Iterator (NoDBBaseConstructor, unsigned short method_name, Model &model)
 alternate constructor for base iterator classes constructed on the fly More...
 
 Iterator (NoDBBaseConstructor, unsigned short method_name)
 alternate constructor for base iterator classes constructed on the fly More...
 
virtual const VariablesArray & initial_points () const
 gets the multiple initial points for this iterator. This will only be meaningful after a call to initial_points mutator.
 
StrStrSizet run_identifier () const
 get the unique run identifier based on method name, id, and number of executions
 

Protected Attributes

Iterator approxSubProbMinimizer
 the minimizer used on the surrogate model to solve the approximate subproblem on each surrogate-based iteration
 
int sbIterNum
 surrogate-based minimization iteration number
 
RealVectorList sbFilter
 Set of response function vectors defining a filter (objective vs. constraint violation) for iterate selection/rejection.
 
RealVector lagrangeMult
 Lagrange multipliers for basic Lagrangian calculations.
 
RealVector augLagrangeMult
 Lagrange multipliers for augmented Lagrangian calculations.
 
Real penaltyParameter
 the penalization factor for violated constraints used in quadratic penalty calculations; increased in update_penalty()
 
RealVector origNonlinIneqLowerBnds
 original nonlinear inequality constraint lower bounds (no relaxation)
 
RealVector origNonlinIneqUpperBnds
 original nonlinear inequality constraint upper bounds (no relaxation)
 
RealVector origNonlinEqTargets
 original nonlinear equality constraint targets (no relaxation)
 
Real eta
 constant used in etaSequence updates
 
Real alphaEta
 power for etaSequence updates when updating penalty
 
Real betaEta
 power for etaSequence updates when updating multipliers
 
Real etaSequence
 decreasing sequence of allowable constraint violation used in augmented Lagrangian updates (refer to Conn, Gould, and Toint, section 14.4)
 
size_t miPLIndex
 index for the active ParallelLevel within ParallelConfiguration::miPLIters
 
- Protected Attributes inherited from Minimizer
size_t numFunctions
 number of response functions
 
size_t numContinuousVars
 number of active continuous vars
 
size_t numDiscreteIntVars
 number of active discrete integer vars
 
size_t numDiscreteStringVars
 number of active discrete string vars
 
size_t numDiscreteRealVars
 number of active discrete real vars
 
Real constraintTol
 optimizer/least squares constraint tolerance
 
Real bigRealBoundSize
 cutoff value for inequality constraint and continuous variable bounds
 
int bigIntBoundSize
 cutoff value for discrete variable bounds
 
size_t numNonlinearIneqConstraints
 number of nonlinear inequality constraints
 
size_t numNonlinearEqConstraints
 number of nonlinear equality constraints
 
size_t numLinearIneqConstraints
 number of linear inequality constraints
 
size_t numLinearEqConstraints
 number of linear equality constraints
 
size_t numNonlinearConstraints
 total number of nonlinear constraints
 
size_t numLinearConstraints
 total number of linear constraints
 
size_t numConstraints
 total number of linear and nonlinear constraints
 
bool optimizationFlag
 flag for use where optimization and NLS must be distinguished
 
size_t numUserPrimaryFns
 number of objective functions or least squares terms in the inbound model; always initialize at Minimizer, even if overridden later
 
size_t numIterPrimaryFns
 number of objective functions or least squares terms in iterator's view, after transformations; always initialize at Minimizer, even if overridden later
 
bool boundConstraintFlag
 convenience flag for denoting the presence of user-specified bound constraints. Used for method selection and error checking.
 
bool speculativeFlag
 flag for speculative gradient evaluations
 
bool calibrationDataFlag
 flag indicating whether user-supplied calibration data is active
 
ExperimentData expData
 Container for experimental data to which to calibrate model using least squares or other formulations which minimize SSE.
 
size_t numExperiments
 number of experiments
 
size_t numTotalCalibTerms
 number of total calibration terms (sum over experiments of number of experimental data per experiment, including field data)
 
Model dataTransformModel
 Shallow copy of the data transformation model, when present (cached in case further wrapped by other transformations)
 
bool scaleFlag
 whether Iterator-level scaling is active
 
Model scalingModel
 Shallow copy of the scaling transformation model, when present (cached in case further wrapped by other transformations)
 
MinimizerprevMinInstance
 pointer containing previous value of minimizerInstance
 
bool vendorNumericalGradFlag
 convenience flag for gradient_type == numerical && method_source == vendor
 
- Protected Attributes inherited from Iterator
ProblemDescDBprobDescDB
 class member reference to the problem description database More...
 
ParallelLibraryparallelLib
 class member reference to the parallel library
 
ParConfigLIter methodPCIter
 the active ParallelConfiguration used by this Iterator instance
 
Model iteratedModel
 the model to be iterated (for iterators and meta-iterators employing a single model instance)
 
size_t myModelLayers
 number of Models locally (in Iterator or derived classes) wrapped around the initially passed in Model
 
unsigned short methodName
 name of the iterator (the user's method spec)
 
Real convergenceTol
 iteration convergence tolerance
 
int maxIterations
 maximum number of iterations for the iterator
 
int maxFunctionEvals
 maximum number of fn evaluations for the iterator
 
int maxEvalConcurrency
 maximum number of concurrent model evaluations More...
 
ActiveSet activeSet
 the response data requirements on each function evaluation
 
size_t numFinalSolutions
 number of solutions to retain in best variables/response arrays
 
VariablesArray bestVariablesArray
 collection of N best solution variables found during the study; always in context of Model originally passed to the Iterator (any in-flight Recasts must be undone)
 
ResponseArray bestResponseArray
 collection of N best solution responses found during the study; always in context of Model originally passed to the Iterator (any in-flight Recasts must be undone)
 
bool subIteratorFlag
 flag indicating if this Iterator is a sub-iterator (NestedModel::subIterator or DataFitSurrModel::daceIterator)
 
SizetArray primaryACVarMapIndices
 "primary" all continuous variable mapping indices flowed down from higher level iteration
 
SizetArray primaryADIVarMapIndices
 "primary" all discrete int variable mapping indices flowed down from higher level iteration
 
SizetArray primaryADSVarMapIndices
 "primary" all discrete string variable mapping indices flowed down from higher level iteration
 
SizetArray primaryADRVarMapIndices
 "primary" all discrete real variable mapping indices flowed down from higher level iteration
 
ShortArray secondaryACVarMapTargets
 "secondary" all continuous variable mapping targets flowed down from higher level iteration
 
ShortArray secondaryADIVarMapTargets
 "secondary" all discrete int variable mapping targets flowed down from higher level iteration
 
ShortArray secondaryADSVarMapTargets
 "secondary" all discrete string variable mapping targets flowed down from higher level iteration
 
ShortArray secondaryADRVarMapTargets
 "secondary" all discrete real variable mapping targets flowed down from higher level iteration
 
short outputLevel
 output verbosity level: {SILENT,QUIET,NORMAL,VERBOSE,DEBUG}_OUTPUT
 
bool summaryOutputFlag
 flag for summary output (evaluation stats, final results); default true, but false for on-the-fly (helper) iterators and sub-iterator use cases
 
ResultsManagerresultsDB
 reference to the global iterator results database
 
ResultsNames resultsNames
 valid names for iterator results
 

Additional Inherited Members

- Public Member Functions inherited from Minimizer
void constraint_tolerance (Real constr_tol)
 set the method constraint tolerance (constraintTol)
 
Real constraint_tolerance () const
 return the method constraint tolerance (constraintTol)
 
bool resize ()
 reinitializes iterator based on new variable size
 
- Static Protected Member Functions inherited from Iterator
static void gnewton_set_recast (const Variables &recast_vars, const ActiveSet &recast_set, ActiveSet &sub_model_set)
 conversion of request vector values for the Gauss-Newton Hessian approximation More...
 
- Static Protected Attributes inherited from Minimizer
static MinimizerminimizerInstance
 pointer to Minimizer used in static member functions
 

Detailed Description

Base class for local/global surrogate-based optimization/least squares.

These minimizers use a SurrogateModel to perform optimization based either on local trust region methods or global updating methods.

Member Function Documentation

void print_results ( std::ostream &  s)
protectedvirtual
void update_lagrange_multipliers ( const RealVector &  fn_vals,
const RealMatrix &  fn_grads 
)
protected

initialize and update Lagrange multipliers for basic Lagrangian

For the Rockafellar augmented Lagrangian, simple Lagrange multiplier updates are available which do not require the active constraint gradients. For the basic Lagrangian, Lagrange multipliers are estimated through solution of a nonnegative linear least squares problem.

References Dakota::abort_handler(), Minimizer::bigRealBoundSize, Minimizer::constraintTol, Iterator::iteratedModel, SurrBasedMinimizer::lagrangeMult, Minimizer::numContinuousVars, Minimizer::numNonlinearEqConstraints, Minimizer::numNonlinearIneqConstraints, Minimizer::numUserPrimaryFns, Minimizer::objective_gradient(), SurrBasedMinimizer::origNonlinIneqLowerBnds, SurrBasedMinimizer::origNonlinIneqUpperBnds, Model::primary_response_fn_sense(), and Model::primary_response_fn_weights().

Referenced by SurrBasedLocalMinimizer::hard_convergence_check().

void update_augmented_lagrange_multipliers ( const RealVector &  fn_vals)
protected

initialize and update the Lagrange multipliers for augmented Lagrangian

For the Rockafellar augmented Lagrangian, simple Lagrange multiplier updates are available which do not require the active constraint gradients. For the basic Lagrangian, Lagrange multipliers are estimated through solution of a nonnegative linear least squares problem.

References SurrBasedMinimizer::augLagrangeMult, SurrBasedMinimizer::betaEta, Minimizer::bigRealBoundSize, SurrBasedMinimizer::etaSequence, Minimizer::numNonlinearEqConstraints, Minimizer::numNonlinearIneqConstraints, Minimizer::numUserPrimaryFns, SurrBasedMinimizer::origNonlinEqTargets, SurrBasedMinimizer::origNonlinIneqLowerBnds, SurrBasedMinimizer::origNonlinIneqUpperBnds, and SurrBasedMinimizer::penaltyParameter.

Referenced by SurrBasedLocalMinimizer::hard_convergence_check(), EffGlobalMinimizer::minimize_surrogates_on_model(), and SurrBasedLocalMinimizer::tr_ratio_check().

bool update_filter ( const RealVector &  fn_vals)
protected
Real lagrangian_merit ( const RealVector &  fn_vals,
const BoolDeque &  sense,
const RealVector &  primary_wts,
const RealVector &  nln_ineq_l_bnds,
const RealVector &  nln_ineq_u_bnds,
const RealVector &  nln_eq_tgts 
)
protected

compute a Lagrangian function from a set of function values

The Lagrangian function computation sums the objective function and the Lagrange multipler terms for inequality/equality constraints. This implementation follows the convention in Vanderplaats with g<=0 and h=0. The bounds/targets passed in may reflect the original constraints or the relaxed constraints.

References Minimizer::bigRealBoundSize, Minimizer::constraintTol, SurrBasedMinimizer::lagrangeMult, Minimizer::numNonlinearEqConstraints, Minimizer::numNonlinearIneqConstraints, Minimizer::numUserPrimaryFns, and Minimizer::objective().

Referenced by SurrBasedLocalMinimizer::approx_subprob_objective_eval(), and SurrBasedLocalMinimizer::tr_ratio_check().

Real augmented_lagrangian_merit ( const RealVector &  fn_vals,
const BoolDeque &  sense,
const RealVector &  primary_wts,
const RealVector &  nln_ineq_l_bnds,
const RealVector &  nln_ineq_u_bnds,
const RealVector &  nln_eq_tgts 
)
protected

compute an augmented Lagrangian function from a set of function values

The Rockafellar augmented Lagrangian function sums the objective function, Lagrange multipler terms for inequality/equality constraints, and quadratic penalty terms for inequality/equality constraints. This implementation follows the convention in Vanderplaats with g<=0 and h=0. The bounds/targets passed in may reflect the original constraints or the relaxed constraints.

References SurrBasedMinimizer::augLagrangeMult, Minimizer::bigRealBoundSize, Minimizer::numNonlinearEqConstraints, Minimizer::numNonlinearIneqConstraints, Minimizer::numUserPrimaryFns, Minimizer::objective(), and SurrBasedMinimizer::penaltyParameter.

Referenced by SurrBasedLocalMinimizer::approx_subprob_objective_eval(), EffGlobalMinimizer::get_best_sample(), EffGlobalMinimizer::minimize_surrogates_on_model(), and SurrBasedLocalMinimizer::tr_ratio_check().

Real penalty_merit ( const RealVector &  fn_vals,
const BoolDeque &  sense,
const RealVector &  primary_wts 
)
protected

compute a penalty function from a set of function values

The penalty function computation applies a quadratic penalty to any constraint violations and adds this to the objective function(s) p = f + r_p cv.

References SurrBasedMinimizer::constraint_violation(), Minimizer::constraintTol, Minimizer::objective(), and SurrBasedMinimizer::penaltyParameter.

Referenced by SurrBasedLocalMinimizer::tr_ratio_check().

Real constraint_violation ( const RealVector &  fn_vals,
const Real &  constraint_tol 
)
protected

compute the constraint violation from a set of function values

Compute the quadratic constraint violation defined as cv = g+^T g+

  • h+^T h+. This implementation supports equality constraints and 2-sided inequalities. The constraint_tol allows for a small constraint infeasibility (used for penalty methods, but not Lagrangian methods).

References Minimizer::bigRealBoundSize, Minimizer::numNonlinearEqConstraints, Minimizer::numNonlinearIneqConstraints, Minimizer::numUserPrimaryFns, SurrBasedMinimizer::origNonlinEqTargets, SurrBasedMinimizer::origNonlinIneqLowerBnds, and SurrBasedMinimizer::origNonlinIneqUpperBnds.

Referenced by SurrBasedLocalMinimizer::hard_convergence_check(), EffGlobalMinimizer::minimize_surrogates_on_model(), SurrBasedMinimizer::penalty_merit(), SurrBasedLocalMinimizer::relax_constraints(), SurrBasedLocalMinimizer::tr_ratio_check(), SurrBasedMinimizer::update_filter(), and SurrBasedLocalMinimizer::update_penalty().


The documentation for this class was generated from the following files: