🤖 AI Summary
This work addresses forward uncertainty propagation in complex physical systems driven by random field inputs by proposing a two-stage mean-variance DeepONet framework that explicitly models the off-diagonal covariance structure among output field variables. By decoupling the learning of output basis functions from the mapping of inputs to expansion coefficients, the method shifts probabilistic modeling from a high-dimensional output space to a low-dimensional orthogonal coefficient subspace, thereby preserving spatial dependencies without explicitly parameterizing the full covariance matrix. Through basis orthonormalization, subspace rotation, and low-rank covariance compression, the model efficiently predicts conditional off-diagonal covariances in a single forward pass. Demonstrated on multiple PDEs and hypersonic aerothermal problems, the approach significantly outperforms Prob-DeepONet in generalization, accurately recovers spatial correlations, and generates structured uncertainty bands.
📝 Abstract
Forward uncertainty propagation in complex physical systems can induce structured covariance across field-valued outputs. For a probabilistic surrogate, the total predictive covariance comprises the covariance of conditional means across input realizations and the average conditional predictive covariance. Probabilistic DeepONet (Prob-DeepONet) provides lightweight uncertainty quantification by predicting pointwise Gaussian means and variances in a single forward pass, but its conditional predictive covariance is restricted to a diagonal form. To represent cross-location conditional dependence without explicitly parameterizing a full high-dimensional covariance matrix, we develop a two-step mean-variance DeepONet (two-step MV-DeepONet) through two principal modifications. First, two-step training is used to decouple output-basis learning from the input-to-coefficient mapping, together with basis orthogonalization and subspace rotation. Second, Gaussian probabilistic modeling is transferred from the high-dimensional physical output space to the low-dimensional rotated coefficient space. Mapping these probabilistic coefficients through the shared basis induces a generally non-diagonal conditional predictive covariance in the physical output space while retaining single-pass inference. A Frobenius-norm error decomposition and corresponding upper bound identify low-rank covariance compressibility, trunk-subspace approximation, finite-sample statistical error, and coefficient-space covariance estimation as the principal factors governing covariance recovery. Numerical experiments on three representative problems governed by partial differential equations (PDEs) and a hypersonic blunt-body aerothermal problem show improved generalization, more structured uncertainty bands, and accurate recovery of off-diagonal correlation patterns compared with Prob-DeepONet.