Target Shapley effects for high-dimensional correlated reliability problems can be estimated from a single failing sample by rewriting closed target Sobol indices via conditional densities and fitting those densities with normalizing flows.
Rare event modeling with self-regularized normalizing flows: what can we learn from a single failure? Preprint:https://arxiv.org/abs/2502.21110
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
High-dimensional reliability-oriented Shapley effect estimation with Normalizing Flows
Target Shapley effects for high-dimensional correlated reliability problems can be estimated from a single failing sample by rewriting closed target Sobol indices via conditional densities and fitting those densities with normalizing flows.