Learning surrogate models for simulation‐based optimization

@article{Cozad2014LearningSM,
  title={Learning surrogate models for simulation‐based optimization},
  author={Alison Cozad and Nikolaos V. Sahinidis and David C. Miller},
  journal={Aiche Journal},
  year={2014},
  volume={60},
  pages={2211-2227}
}
A central problem in modeling, namely that of learning an algebraic model from data obtained from simulations or experiments is addressed. A methodology that uses a small number of simulations or experiments to learn models that are as accurate and as simple as possible is proposed. The approach begins by building a low-complexity surrogate model. The model is built using a best subset technique that leverages an integer programming formulation to allow for the efficient consideration of a… 
Surrogate-based optimization for mixed-integer nonlinear problems
TLDR
This work proposes an algorithm for mixed-integer nonlinear simulation-based problems that uses adaptive sampling and surrogate modeling with one-hot encoding and results show that one- hot encoding leads to accurate and robust mixed-variable Gaussian Process and Neural Network models that are effective surrogates for optimization.
Machine learning-based surrogate modeling for data-driven optimization: a comparison of subset selection for regression techniques
TLDR
This work investigates the efficiency of subset selection regression techniques for developing surrogate functions that balance both accuracy and complexity and indicates that subset selection-based regression functions exhibit promising performance when the dimensionality is low, while interpolation performs better for higher dimensional problems.
Managing uncertainty in data-driven simulation-based optimization
TLDR
It is shown that incorporating uncertainty management within simulation-based optimization leads to more robust solutions, which protect the decision-maker from infeasible solutions.
The ALAMO approach to machine learning
TLDR
ALAMO's constrained regression methodology is used to further refine concentration models, resulting in models that perform better on validation data and satisfy upper and lower bounds placed on model outputs.
A Genetic Programming Approach for Construction of Surrogate Models
TLDR
A methodology that combines iterative Design of experiments (DOE) with Genetic Programming (GP) in order to obtain surrogate models is presented and results indicate that the proposed methodology is able to provide accurate surrogate models for the variables.
Self-learning surrogate models in superstructure optimization
TLDR
Initial results indicate that the suggested methodology holds potential to improve computational time and flexibility and the methodology of continuously adding the data points based on the prediction uncertainty improves the quality of the surrogate model.
Advances in surrogate based modeling, feasibility analysis, and optimization: A review
TLDR
Two of the frequently used surrogates, radial basis functions, and Kriging are tested on a variety of test problems and guidelines for the choice of appropriate surrogate model are discussed.
Surrogate Model Selection for Design Space Approximation And Surrogatebased Optimization
TLDR
Evaluating the performance of eight surrogate modeling techniques for design space approximation and surrogate based optimization applications over a set of generated datasets with known characteristics revealed that multivariate adaptive regression spline models and single hidden layer feed forward neural networks yielded the most accurate predictions over the design space.
A Trust Region Filter Algorithm for Surrogate-based Optimization
TLDR
This work proposes an algorithm that combines ideas from SQP filter methods and derivative free trust region methods to solve the general problem of hybrid glass box/black box optimization, with focus on guaranteeing that a surrogate-based optimization strategy converges to optimal points of the original detailed model.
Data-Driven Spatial Branch-And-Bound Algorithms For Black-Box Optimization
TLDR
A novel data-driven spatial branch-and-bound framework is proposed that uses stochastic bounds on different surrogate models and partitioning of the search space to mitigate uncertainty in the performance of surrogate-based optimization algorithms.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 47 REFERENCES
A Surrogate Modeling and Adaptive Sampling Toolbox for Computer Based Design
TLDR
This paper presents a mature, flexible, and adaptive machine learning toolkit for regression modeling and active learning to tackle issues of computational cost and model accuracy.
A novel sequential design strategy for global surrogate modeling
TLDR
A comparison is made between different sequential design methods for global surrogate modeling on a real-world electronics problem using a novel hybrid technique which incorporates both an exploitation criterion, using local linear approximations of the objective function, and an exploration criteria, using a Monte Carlo Voronoi tessellation.
Efficient space-filling and non-collapsing sequential design strategies for simulation-based modeling
TLDR
An extensive study of new and state-of-the-art space-filling sequential design methods is performed, and it is shown that the new sequential methods proposed are comparable to the best one-shot experimental designs available right now.
An algorithm for the use of surrogate models in modular flowsheet optimization
In this work a methodology is presented for the rigorous optimization of nonlinear programming problems in which the objective function and (or) some constraints are represented by noisy implicit
A kriging method for the solution of nonlinear programs with black‐box functions
In this article, a new methodology is developed for the optimization of black-box systems lacking a closed-form mathematical description. To properly balance the computational cost of building the
Efficient Global Optimization of Expensive Black-Box Functions
TLDR
This paper introduces the reader to a response surface methodology that is especially good at modeling the nonlinear, multimodal functions that often occur in engineering and shows how these approximating functions can be used to construct an efficient global optimization algorithm with a credible stopping rule.
Surrogate‐based superstructure optimization framework
In principle, optimization-based “superstructure” methods for process synthesis can be more powerful than sequential-conceptual methods as they account for all complex interactions between design
Global Optimization of Stochastic Black-Box Systems via Sequential Kriging Meta-Models
TLDR
The proposed method is based on a kriging meta-model that provides a global prediction of the objective values and a measure of prediction uncertainty at every point and has excellent consistency and efficiency in finding global optimal solutions.
Optimization for Simulation: Theory vs. Practice
  • M. Fu
  • Computer Science
  • 2002
TLDR
There is a disconnect between research in simulation optimization—which has addressed the stochastic nature of discrete-event simulation by concentrating on theoretical results of convergence and specialized algorithms that are mathematically elegant—and the recent software developments, which implement very general algorithms adopted from techniques in the deterministic optimization metaheuristic literature.
Feature Article: Optimization for simulation: Theory vs. Practice
  • M. Fu
  • Computer Science
    INFORMS J. Comput.
  • 2002
TLDR
There is a disconnect between research in simulation optimization--which has addressed the stochastic nature of discrete-event simulation by concentrating on theoretical results of convergence and specialized algorithms that are mathematically elegant--and the recent software developments, which implement very general algorithms adopted from techniques in the deterministic optimization metaheuristic literature.
...
1
2
3
4
5
...