Sequential active learning of low-dimensional model representations for reliability analysis

DOI10.1137/21M1416758zbMATH Open1493.62496arXiv2106.04453OpenAlexW3169986546MaRDI QIDQ5864079FDOQ5864079

Authors: Max Ehre, Iason Papaioannou, Bruno Sudret, Daniel Straub 

Publication date: 3 June 2022

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Abstract: To date, the analysis of high-dimensional, computationally expensive engineering models remains a difficult challenge in risk and reliability engineering. We use a combination of dimensionality reduction and surrogate modelling termed partial least squares-driven polynomial chaos expansion (PLS-PCE) to render such problems feasible. Standalone surrogate models typically perform poorly for reliability analysis. Therefore, in a previous work, we have used PLS-PCEs to reconstruct the intermediate densities of a sequential importance sampling approach to reliability analysis. Here, we extend this approach with an active learning procedure that allows for improved error control at each importance sampling level. To this end, we formulate an estimate of the combined estimation error for both the subspace identified in the dimension reduction step and surrogate model constructed therein. With this, it is possible to adapt the design of experiments so as to optimally learn the subspace representation and the surrogate model constructed therein. The approach is gradient-free and thus can be directly applied to black box-type models. We demonstrate the performance of this approach with a series of low- (2 dimensions) to high- (869 dimensions) dimensional example problems featuring a number of well-known caveats for reliability methods besides high dimensions and expensive computational models: strongly nonlinear limit-state functions, multiple relevant failure regions and small probabilities of failure.

Full work available at URL: https://arxiv.org/abs/2106.04453

Recommendations

• Surrogate assisted active subspace and active subspace assisted surrogate -- a new paradigm for high dimensional structural reliability analysis
• Active learning relevant vector machine for reliability analysis
• PLS-based adaptation for efficient PCE representation in high dimensions
• Efficient reliability analysis with a CDA-based dimension-reduction model and polynomial chaos expansion
• Efficient reliability analysis based on adaptive sequential sampling design and cross-validation

zbMATH Keywords

dimensionality reductionsequential importance samplingrare event simulationactive learningreliability analysisPLS-PCE

Mathematics Subject Classification ID

General nonlinear regression (62J02) Applications of statistics in engineering and industry; control charts (62P30) Monte Carlo methods (65C05) Sequential statistical methods (62L99)

Cites Work

• Quasi-regression
• 10.1162/15324430152748236
• The Collinearity Problem in Linear Regression. The Partial Least Squares (PLS) Approach to Generalized Inverses
• Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit
• Simulation and the Monte Carlo Method
• The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
• Asymptotic theory of nonlinear least squares estimation
• Stochastic collocation algorithms using \(\ell_1\)-minimization
• A non-adapted sparse approximation of PDEs with stochastic inputs
• Title not available (Why is that?)
• Bayesian Compressive Sensing
• Adaptive sparse polynomial chaos expansion based on least angle regression
• Sparse polynomial chaos expansions and adaptive stochastic finite elements using a regression approach
• Basis adaptation in homogeneous chaos spaces
• Stochastic finite element: a non intrusive approach by regression
• Sparse polynomial chaos expansions: literature survey and benchmark
• Dimensionality reduction for complex models via Bayesian compressive sensing
• Structural reliability under combined random load sequences
• An efficient surrogate-based method for computing rare failure probability
• Evaluation of failure probability via surrogate models
• Multilevel sequential importance sampling for rare event estimation
• Model selection for small sample regression
• Sparse polynomial chaos expansions using variational relevance vector machines
• Structural reliability analysis of elastic-plastic structures using neural networks and Monte Carlo simulation
• Estimating Percentiles of Uncertain Computer Code Outputs
• Neural-network-based realibility analysis: A comparative study
• Accelerated subset simulation with neural networks for reliability analysis
• Multilevel estimation of rare events
• Erratum: Active Subspace Methods in Theory and Practice: Applications to Kriging Surfaces
• Polynomial meta-models with canonical low-rank approximations: numerical insights and comparison to sparse polynomial chaos expansions
• Compressive sensing adaptation for polynomial chaos expansions
• PLS-based adaptation for efficient PCE representation in high dimensions
• Uncertainty propagation of p-boxes using sparse polynomial chaos expansions
• Bayesian Subset Simulation
• Cross-Entropy-Based Importance Sampling with Failure-Informed Dimension Reduction for Rare Event Simulation
• Multifidelity preconditioning of the cross-entropy method for rare event simulation and failure probability estimation

Cited In (6)

• Active learning polynomial chaos expansion for reliability analysis by maximizing expected indicator function prediction error
• PLS-based adaptation for efficient PCE representation in high dimensions
• Active-subspace analysis of exceedance probability for shallow-water waves
• A novel active learning method based on matrix-operation RBF model for high-dimensional reliability analysis
• A survey of unsupervised learning methods for high-dimensional uncertainty quantification in black-box-type problems
• Surrogate assisted active subspace and active subspace assisted surrogate -- a new paradigm for high dimensional structural reliability analysis

Uses Software

• AK-MCS

This page was built for publication: Sequential active learning of low-dimensional model representations for reliability analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5864079)