A new surrogate refinement strategy guided by adaptive importance sampling proposals achieves comparable rare-event accuracy to full-model methods while using far fewer high-fidelity PDE evaluations in up to 100 dimensions.
Deep uq: Learning deep neural network surrogate models for high dimensional uncertainty quantification.Journal of computational physics, 375:565– 588, 2018
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Proposal-Guided Greedy Surrogate Refinement for PDE-Driven High-Dimensional Rare-Event Estimation
A new surrogate refinement strategy guided by adaptive importance sampling proposals achieves comparable rare-event accuracy to full-model methods while using far fewer high-fidelity PDE evaluations in up to 100 dimensions.