A trust-region method for derivative-free nonlinear constrained stochastic optimization

12 Mar 2017  ·  Augustin F., Marzouk Y. M. ·

In this work we introduce the algorithm (S)NOWPAC (Stochastic Nonlinear Optimization With Path-Augmented Constraints) for stochastic nonlinear constrained derivative-free optimization. The algorithm extends the derivative-free optimizer NOWPAC to be applicable to nonlinear stochastic programming... It is based on a trust region framework, utilizing local fully linear surrogate models combined with Gaussian process surrogates to mitigate the noise in the objective function and constraint evaluations. We show several benchmark results that demonstrate (S)NOWPAC's efficiency and highlight the accuracy of the optimal solutions found. read more

PDF Abstract
No code implementations yet. Submit your code now


Optimization and Control