Quantifying uncertainty with ensembles of surrogates for blackbox optimization
From MaRDI portal
Publication:2162525
DOI10.1007/S10589-022-00381-ZzbMATH Open1496.90091arXiv2107.04360OpenAlexW3180164923MaRDI QIDQ2162525FDOQ2162525
Authors: Charles Audet, Sébastien Le Digabel, Renaud Saltet 
Publication date: 8 August 2022
Published in: Computational Optimization and Applications (Search for Journal in Brave)
Abstract: This work is in the context of blackbox optimization where the functions defining the problem are expensive to evaluate and where no derivatives are available. A tried and tested technique is to build surrogates of the objective and the constraints in order to conduct the optimization at a cheaper computational cost. This work proposes different uncertainty measures when using ensembles of surrogates. The resulting combination of an ensemble of surrogates with our measures behaves as a stochastic model and allows the use of efficient Bayesian optimization tools. The method is incorporated in the search step of the mesh adaptive direct search (MADS) algorithm to improve the exploration of the search space. Computational experiments are conducted on seven analytical problems, two multi-disciplinary optimization problems and two simulation problems. The results show that the proposed approach solves expensive simulation-based problems at a greater precision and with a lower computational effort than stochastic models.
Full work available at URL: https://arxiv.org/abs/2107.04360
Recommendations
- Surrogate‐based methods for black‐box optimization
- High-dimensional black-box optimization under uncertainty
- A survey on direct search methods for blackbox optimization and their applications
- Influence of ensemble surrogate models and sampling strategy on the solution quality of algorithms for~computationally expensive black-box global optimization problems
- The mesh adaptive direct search algorithm with treed Gaussian process surrogates
derivative-free optimizationmesh adaptive direct searchblackbox optimizationBayesian optimizationensembles of surrogates
Nonlinear programming (90C30) Derivative-free methods and methods using generalized derivatives (90C56)
Cites Work
- Efficient global optimization of expensive black-box functions
- Algorithm 1027: NOMAD Version 4: Nonlinear Optimization with the MADS Algorithm
- A Linesearch-Based Derivative-Free Approach for Nonsmooth Constrained Optimization
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- Gaussian processes for machine learning.
- Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit Setting
- A taxonomy of global optimization methods based on response surfaces
- A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code
- Optimization and nonsmooth analysis
- A progressive barrier for derivative-free nonlinear programming
- Mesh Adaptive Direct Search Algorithms for Constrained Optimization
- Test examples for nonlinear programming codes
- On the Convergence of Pattern Search Algorithms
- Introduction to Derivative-Free Optimization
- Constrained global optimization of expensive black box functions using radial basis functions
- Finding Optimal Algorithmic Parameters Using Derivative‐Free Optimization
- Theory of Positive Linear Dependence
- Benchmarking Derivative-Free Optimization Algorithms
- Title not available (Why is that?)
- Title not available (Why is that?)
- Global optimization of costly nonconvex functions using radial basis functions
- Optimizing radial basis functions by d.c. programming and its use in direct search for global derivative-free optimization
- Efficient global optimization algorithm assisted by multiple surrogate techniques
- Parallel Space Decomposition of the Mesh Adaptive Direct Search Algorithm
- Trailing-edge noise reduction using derivative-free optimization and large-eddy simulation
- OrthoMADS: A Deterministic MADS Instance with Orthogonal Directions
- Use of quadratic models with mesh-adaptive direct search for constrained black box optimization
- Stochastic radial basis function algorithms for large-scale optimization involving expensive black-box objective and constraint functions
- Spent potliner treatment process optimization using a MADS algorithm
- Nonsmooth optimization through mesh adaptive direct search and variable neighborhood search
- Mixture surrogate models based on Dempster-Shafer theory for global optimization problems
- Detection and Remediation of Stagnation in the Nelder--Mead Algorithm Using a Sufficient Decrease Condition
- ORBIT: Optimization by Radial Basis Function Interpolation in Trust-Regions
- Surrogate Optimization of Computationally Expensive Black-Box Problems with Hidden Constraints
- Surrogate‐based methods for black‐box optimization
- An algorithmic framework for the optimization of computationally expensive bi-fidelity black-box problems
- Dynamic trees for learning and design
- Incorporating minimum Frobenius norm models in direct search
- Convergence results for generalized pattern search algorithms are tight
- The mesh adaptive direct search algorithm with treed Gaussian process surrogates
- Global convergence of radial basis function trust region derivative-free algorithms
- Surrogate optimization of deep neural networks for groundwater predictions
- Order-based error for managing ensembles of surrogates in mesh adaptive direct search
- Locally weighted regression models for surrogate-assisted design optimization
- Derivative-free and blackbox optimization
- On the construction of quadratic models for derivative-free trust-region algorithms
- Derivative-free optimization methods
- Optimization of automotive valve train components with implicit filtering
- Sequential approximate optimization using radial basis function network for engineering optimization
- Dynamic improvements of static surrogates in direct search optimization
Cited In (11)
- Order-based error for managing ensembles of surrogates in mesh adaptive direct search
- Surrogate-based branch-and-bound algorithms for simulation-based black-box optimization
- High-dimensional black-box optimization under uncertainty
- Influence of ensemble surrogate models and sampling strategy on the solution quality of algorithms for~computationally expensive black-box global optimization problems
- Surrogate-based feasibility analysis for black-box stochastic simulations with heteroscedastic noise
- A survey of unsupervised learning methods for high-dimensional uncertainty quantification in black-box-type problems
- Title not available (Why is that?)
- Surrogate-Based Ensemble Grouping Strategies for Embedded Sampling-Based Uncertainty Quantification
- Black-box combinatorial optimization using models with integer-valued minima
- Surrogate-based distributed optimisation for expensive black-box functions
- Surrogate‐based methods for black‐box optimization
Uses Software
This page was built for publication: Quantifying uncertainty with ensembles of surrogates for blackbox optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2162525)
