A rigorous framework for optimization of expensive functions by surrogates

By Andrew, Booker, Paul, Frankt, David, Serafini, Virginia, Torczon. 1998

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Accessorize to a Crime: Real and Stealthy Attacks on State-of-the-Art Face Recognition (2016, methodology): “To ﬁnd the perturbation, the parameters κ 1 and κ 2 in the optimization were set to 0.15 and 0.25, respectively.”
Robots that can adapt like animals (2014, methodology): “Like all model-based optimization algorithms (e.g. surrogate-based algorithms 36,37,38 , kriging 39 , or DACE 40,41 ), Bayesian optimization creates a model of the objective function with a regression method, uses this model to select the next point to acquire, then updates the model, etc.”
Optimal Model Management for Multifidelity Monte Carlo Estimation (2016, background): “models to accelerate convergence to an optimal solution [4, 1, 30, 21].”
A comparison of deterministic and probabilistic optimization algorithms for nonsmooth simulation-based optimization (2004, methodology): “Other promising methods that have successfully been used in simulation-based optimization, such as methods that use surrogate models [16], are beyond the scope of this paper.”
Surrogate-based Analysis and Optimization (2005, background): “Being based on pattern search methods, the SMF framework: (i) can be used for scenarios where the objective functions are nondifferentiable, or for which sensitivities are difficult or expensive to attain, (ii) can be easily adapted to either a parallel or distributed computing environment, (iii) is”
Comparative studies of metamodeling techniques under multiple modeling criteria (2000, methodology): “An interpolation method known as Kriging is becoming widely used for the design and analysis of computer experiments (Sacks, et al., 1989; Booker, et al., 1999 ).”
Provably Convergent Multifidelity Optimization Algorithm Not Requiring High-Fidelity Derivatives (2012, methodology): “management framework (SMF) [10] can be used as they do not require high-fidelity gradient estimates.”
On Sequential Sampling for Global Metamodeling in Engineering Design (2002, background): “For metamodel-based design optimization, based on different criteria for a balanced search (sequential sampling), various optimization approaches have been developed, e.g., Efficient Global Optimization (EGO) [17], Surrogate Management Framework (SMF) [5], Sequential Design Optimization (SDO) [18], ”
A Tutorial on Bayesian Optimization (2018, methodology): “While we do not review methods from this literature here in detail, many of them have a similar ﬂavor to BayesOpt methods: they maintain a surrogate that models the objective function, which they use to choose where to evaluate (Booker et al., 1999; Regis and Shoemaker, 2007b,a, 2005).”
Metamodel based high-fidelity stochastic analysis of composite laminates: A concise review with critical comparative assessment (2017, methodology): “An interpolation method known as Kriging is widely utilised for the design and analysis of computer experiments [12-14].”
A Provably Convergent Multifidelity Optimization Algorithm not Requiring High-Fidelity Derivatives (2010, methodology): “Considering shock-expansion theory as the high-fidelity analysis and the panel method as the lowfidelity analysis, Table 4 compares the average number of highfidelity evaluations from random initial airfoils parametrized by the angle of attack and ten surface spline points.”