Parallel surrogate-assisted optimization: Batched Bayesian Neural Network-assisted GA versus q-EGO

Abstract

Surrogate-based optimization is widely used to deal with long-running black-box simulation-based objective functions. Actually, the use of a surrogate model such as Kriging or Artificial Neural Network allows to reduce the number of calls to the CPU time-intensive simulator. Bayesian optimization uses the ability of surrogates to provide useful information to help guiding effectively the optimization process. In this paper, the Efficient Global Optimization (EGO) reference framework is challenged by a Bayesian Neural Network-assisted Genetic Algorithm, namely BNN-GA. The Bayesian Neural Network (BNN) surrogate is chosen for its ability to provide an uncertainty measure of the prediction that allows to compute the Expected Improvement of a candidate solution in order to improve the exploration of the objective space. BNN is also more reliable than Kriging models for high-dimensional problems and faster to set up thanks to its incremental training. In addition, we propose a batch-based approach for the parallelization of BNN-GA that is challenged by a parallel version of EGO, called q-EGO. Parallel computing is a highly important complementary way (to surrogates) to deal with the computational burden of simulation-based optimization. The comparison of the two parallel approaches is experimentally performed through several benchmark functions and two real-world problems within the scope of Tuberculosis Transmission Control (TBTC). The study presented in this paper proves that parallel batched BNN-GA is a viable alternative to q-EGO approaches being more suitable for high-dimensional problems, parallelization impact, bigger databases and moderate search budgets. Moreover, a significant improvement of the solutions is obtained for the two TBTC problems tackled.

Speeding up Computational Morphogenesis with Online Neural Synthetic Gradients

“…More recently, attempts have been made to apply machine learning to accelerate PDE-constrained optimization problems. The surrogate-assisted optimization approaches [32], which use various machine learning models such as Gaussian process [33] and deep neural networks [34] to replace expensive function evaluations, are popular to speed up gradient-free optimization problems. However, those approaches are not suitable for topology optimization problems which typically require gradient-based optimizers [35].…”

A Trust Index-Based Adaptive Kriging Method for the Cooling Fan Optimization

“…Therefore, many studies are carried out to improve EI by balance global and local search capabilities [8][9][10]. Meanwhile, since the intelligent algorithm has good exploration ability, they are combined with the surrogate model-based method to form the classic optimization framework [11][12][13]. By constructing fidelity surrogate models with the appropriate intelligent algorithm, the design cycle of the product is dramatically shortened.…”

Vortex-model-based Multi-objective Optimization of Winglets for Wind Turbines using Machine Learning

“…Surrogate-assisted modeling can hence reduce the number of iterations needed to reach an optimized design. Machine learning methods such as Bayesian Neural Networks [8] or Gaussian Processes in Bayesian Optimization [9] have recently shown promising results in surrogate-assisted optimization.…”