Maximizing Reliability with Bayesian Optimization
Not on the map. Why?
Abstract (arXiv)
Bayesian optimization (BO) is a popular, sample-efficient technique for expensive, black-box optimization. One such problem arising in manufacturing is that of maximizing the reliability, or equivalently minimizing the probability of a failure, of a design which is subject to random perturbations - a problem that can involve extremely rare failures (Pfail = 10-6-10-8). In this work, we propose two BO methods based on Thompson sampling and knowledge gradient, the latter approximating the one-step Bayes-optimal policy for minimizing the logarithm of the failure probability. Both methods incorporate importance sampling to target extremely small failure probabilities. Empirical results show the proposed methods outperform existing methods in both extreme and non-extreme regimes.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 17 of its 52 references that are in the map. First posted on arXiv 2 Feb 2026.
Builds on
17 of 52 references.
References · 17
- 2020Efficiently Sampling Functions from Gaussian Process Posteriorscited by 125
- 2020Pathwise Conditioning of Gaussian Processescited by 63
- 2022Robust Multi-Objective Bayesian Optimization Under Input Noisecited by 36
- 2020Bayesian Optimization of Risk Measurescited by 35
- 2021Value-at-Risk Optimization with Gaussian Processescited by 19
- 2020Bayesian Quantile and Expectile Optimisationcited by 17
- 2023Bayesian Optimization of Function Networks with Partial Evaluationscited by 10
- 2025Exploring Exploration in Bayesian Optimizationcited by 8
- 2022A Robust Multi-Objective Bayesian Optimization Framework Considering Input Uncertaintycited by 7
- 2023Knowledge Gradient for Multi-Objective Bayesian Optimization with Decoupled Evaluationscited by 5
- 2020Hamiltonian MCMC methods for estimating rare events probabilities in high-dimensional problemscited by 4
- 2022Large deviation theory-based adaptive importance sampling for rare events in high dimensionscited by 4
5 more
- 2023Maximizing the Success Probability of Policy Allocations in Online Systemscited by 4
- 2021A sampling criterion for constrained Bayesian optimization with uncertaintiescited by 3
- 2024Two-stage Design for Failure Probability Estimation with Gaussian Process Surrogatescited by 2
- 2022Coupling and selecting constraints in Bayesian optimization under uncertaintiescited by 1
- 2024Bayesian Optimization for Non-Convex Two-Stage Stochastic Optimization Problemscited by 1
Built on by
Not cited yet.