Variational Bayesian strategies for high-dimensional, stochastic design problems
From MaRDI portal
Abstract: This paper is concerned with a lesser-studied problem in the context of model-based, uncertainty quantification (UQ), that of optimization/design/control under uncertainty. The solution of such problems is hindered not only by the usual difficulties encountered in UQ tasks (e.g. the high computational cost of each forward simulation, the large number of random variables) but also by the need to solve a nonlinear optimization problem involving large numbers of design variables and potentially constraints. We propose a framework that is suitable for a large class of such problems and is based on the idea of recasting them as probabilistic inference tasks. To that end, we propose a Variational Bayesian (VB) formulation and an iterative VB-Expectation-Maximization scheme that is also capable of identifying a low-dimensional set of directions in the design space, along which, the objective exhibits the largest sensitivity. We demonstrate the validity of the proposed approach in the context of two numerical examples involving random and design variables. In all cases considered the cost of the computations in terms of calls to the forward model was of the order . The accuracy of the approximations provided is assessed by appropriate information-theoretic metrics.
Recommendations
- A Bayesian approach for quantile optimization problems with high-dimensional uncertainty sources
- Bayesian optimization of variable-size design space problems
- Variational Bayesian approximation of inverse problems using sparse precision matrices
- Adaptive importance sampling for optimization under uncertainty problems
- Stein variational reduced basis Bayesian inversion
Cites work
- scientific article; zbMATH DE number 3423418 (Why is no real title available?)
- scientific article; zbMATH DE number 3886886 (Why is no real title available?)
- scientific article; zbMATH DE number 107482 (Why is no real title available?)
- scientific article; zbMATH DE number 3513115 (Why is no real title available?)
- scientific article; zbMATH DE number 3567782 (Why is no real title available?)
- scientific article; zbMATH DE number 1222290 (Why is no real title available?)
- scientific article; zbMATH DE number 1972910 (Why is no real title available?)
- scientific article; zbMATH DE number 2038815 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- A Stochastic Approximation Method
- A feasible method for optimization with orthogonality constraints
- A method for stochastic constrained optimization using derivative-free surrogate pattern search and collocation
- A new level-set based approach to shape and topology optimization under geometric uncertainty
- A scalable framework for the solution of stochastic inverse problems using a sparse grid collocation approach
- Accelerating PDE constrained optimization by the reduced basis method: application to batch chromatography
- Advances in neural information processing systems 19. Proceedings of the 2006 conference, Vancouver, BC, Canada, December 4--6, 2006
- An introduction to variational methods for graphical models
- Bayes Factors
- Bayesian Orthogonal Component Analysis for Sparse Representation
- Bayesian thinking, modeling and computation.
- Bayesian-Optimal Design via Interacting Particle Systems
- Certified PDE-constrained parameter optimization using reduced basis surrogate models for evolution problems
- Convergence of a stochastic approximation version of the EM algorithm
- GRADIENT-BASED STOCHASTIC OPTIMIZATION METHODS IN BAYESIAN EXPERIMENTAL DESIGN
- Gaussian Kullback-Leibler approximate inference
- Graphical models, exponential families, and variational inference
- Introduction to Stochastic Search and Optimization
- Large-Scale PDE-Constrained Optimization: An Introduction
- Linear programming under uncertainty
- Multi-objective design optimization using cascade evolutionary computations
- Numerical Optimization
- Optimization with PDE Constraints
- Probabilistic Principal Component Analysis
- Reliability-based structural optimization using neural networks and Monte Carlo simulation
- Robust optimization - a comprehensive survey
- Robust topology optimization accounting for spatially varying manufacturing errors
- Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
- Shape Optimization Under Uncertainty—A Stochastic Programming Perspective
- Simulating normalizing constants: From importance sampling to bridge sampling to path sampling
- Sparse Bayesian Learning for Basis Selection
- Statistical and computational inverse problems.
- Stochastic Estimation of the Maximum of a Regression Function
- Structural optimization under uncertain loads and nodal locations
- Topology optimization of continuum structures under uncertainty - A polynomial chaos approach
- Topology optimization with geometric uncertainties by perturbation techniques
- Two-Point Step Size Gradient Methods
- Variational Mixture of Bayesian Independent Component Analyzers
Cited in
(8)- Generalized Bayes approach to inverse problems with model misspecification
- Collaborative and adaptive Bayesian optimization for bounding variances and probabilities under hybrid uncertainties
- Ensemble Kalman filters and geometric characterization of sensitivity spaces for uncertainty quantification in optimization
- Bayesian optimization of variable-size design space problems
- A Bayesian approach for quantile optimization problems with high-dimensional uncertainty sources
- Special issue: Big data and predictive computational modeling
- scientific article; zbMATH DE number 7733441 (Why is no real title available?)
- Variational Bayesian surrogate modelling with application to robust design optimisation
This page was built for publication: Variational Bayesian strategies for high-dimensional, stochastic design problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q729450)
