Search papers, labs, and topics across Lattice.
This paper introduces a method for analytically extracting conditional Sobol' indices from pre-trained global Polynomial Chaos Expansion (PCE) models, enabling efficient sensitivity analysis for systems with parameterized responses. By exploiting the tensor-product property of PCE bases, the global expansion is reformulated into analytical coefficient fields dependent on conditioning variables. The method derives closed-form expressions for conditional variances and Sobol' indices, eliminating the need for repetitive modeling or additional sampling.
Forget retraining: conditional sensitivity analysis can be reduced to algebraic post-processing of existing Polynomial Chaos Expansion models.
In uncertainty quantification, evaluating sensitivity measures under specific conditions (i.e., conditional Sobol'indices) is essential for systems with parameterized responses, such as spatial fields or varying operating conditions. Traditional approaches often rely on point-wise modeling, which is computationally expensive and may lack consistency across the parameter space. This paper demonstrates that for a pre-trained global Polynomial Chaos Expansion (PCE) model, the analytical conditional Sobol'indices are inherently embedded within its basis functions. By leveraging the tensor-product property of PCE bases, we reformulate the global expansion into a set of analytical coefficient fields that depend on the conditioning variables. Based on the preservation of orthogonality under conditional probability measures, we derive closed-form expressions for conditional variances and Sobol'indices. This framework bypasses the need for repetitive modeling or additional sampling, transforming conditional sensitivity analysis into a purely algebraic post-processing step. Numerical benchmarks indicate that the proposed method ensures physical coherence and offers superior numerical robustness and computational efficiency compared to conventional point-wise approaches.
