Flexibility and efficiency enhancements for constrained global design optimization with kriging approximations.

By M. Sasena. 2002

431 citations recorded by the source corpus.

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

Taking the Human Out of the Loop: A Review of Bayesian Optimization (2016, background): “Since EGO’s publication, there has evolved a body of work devoted to extending the algorithm, particularly in adding constraints to the optimization problem [6], [131], [23], and in modelling noisy functions [14], [75], [76].”
A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning (2010, methodology): “Bayesian optimization techniques are some of the most efficient approaches in terms of the number of function evaluations required (see, e.g. [Močkus, 1994, Jones et al., 1998, Streltsov and Vakili, 1999, Jones, 2001, Sasena, 2002]).”
A Bayesian exploration-exploitation approach for optimal online sensing and planning with a visually guided mobile robot (2009, methodology): “Several refinements of this infill function have been proposed in Schonlau et al (1998) and Sasena (2002).”
Recent Advances in Bayesian Optimization (2022, background): “Sasena [158] gave a review of early work on Kriging and its extension to constrained optimization.”
Gaussian processes for time-series modelling (2013, background): “Perhaps the simplest approach is to take a covariance function that is the product of one-dimensional covariances over each input (the product correlation rule [5]),”
Bayesian Gaussian processes for sequential prediction, optimisation and quadrature (2010, background): “Perhaps the simplest approach is to take a covariance function that is the product of one-dimensional covariances over each input (the product correlation rule [Sasena, 2002]), K(x1, x2; φ) = ∏”
Active Policy Learning for Robot Planning and Exploration under Uncertainty (2007, unknown): “…Bayesian analysis allow us to obtain analytical expressions for the posterior modes of these quantities: µ̂ = (1T K−11 + δ−2)−11T K−1Cπ σ̂2 = b + CπT K−1Cπ − (1T K−11 + δ−2)µ̂2 n + a + 2 Using the previous estimates, the GP predictive mean and variance are given by Ĉπ(θ) = µ̂ + kT K−1(Cπ1:n…”
Infill sampling criteria for surrogate-based optimization with constraint handling (2012, background): “If there is a tolerance associated with satisfying the equality constraints, an equality can be considered as two closely bound inequality constraints (Sasena 2002).”
A benchmark of kriging-based infill criteria for noisy optimization (2013, unknown): “In this case, R̃ij = 1 if i = j, and (1 − ν)r(xi,xj) otherwise, where τ(2) is commonly called nugget and ν = τ 2 σ(2)+τ(2) scaling factor [34].”
A Bayesian interactive optimization approach to procedural animation design (2010, methodology): “5 Experiments The Bayesian optimization approach has been found to be very ef.cient for expensive objective functions [JSW98, Sas02,Liz08], which suggests it is suitable to our task.”
A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design (2017, methodology): “To address this issue, the second switch strategy (Sasena et al. 2002; Sasena 2002) or the greedy strategy (Singh et al. 2013) is proposed, as shown in Fig.”
Use of Kriging Models to Approximate Deterministic Computer Models (2005, methodology): “[9] improved the matrix condition number by adding a small value (10) to the diagonal elements of the correlation matrix, and Sasena [36] used a small nugget effect (e.”
A Tutorial on Bayesian Optimization (2018, background): “Other tutorials and surveys on Bayesian optimization include Shahriari et al. (2016); Brochu et al. (2009); Sasena (2002); Frazier and Wang (2016).”
The Knowledge-Gradient Policy for Correlated Normal Beliefs (2009, background): “Bayesian global optimization, which began with Kushner (1964) and was recently reviewed in Sasena (2002) and Kleijnen (2009), uses a Gaussian process prior to model belief about an unknown function, and then chooses experiments to most efficiently optimize that function.”
Sequential kriging optimization using multiple-fidelity evaluations (2006, methodology): “We assume that the responses of the real system derives from a test function created by Sasena (2002) 10)100/exp()sin()(2 +−−= xxxf (0 x 10) (16) and we arbitrarily assume that the output of the surrogate system is 21 )3(03.03.10)100/exp()sin()( −×++−−= xxxxf (0 x 10).”
Metamodeling in Multidisciplinary Design Optimization: How Far Have We Really Come? (2014, unknown): “dissertations [80–83].”
Kriging metamodeling in simulation: A review (2009, methodology): “• Reference [21] derives sequential designs for the optimization of random and deterministic simulation models, using Kriging and so-called Efficient Global Optimization (EGO), which maximizes the Expected Improvement (EI) following a Bayesian approach; also see [13,22,24,45].”