cs.LGNov 7, 2026

Applied bifidelity material response model construction and uncertainty quantification with a sampling hypernetwork

Authors: Reese E. Jones, Cosmin Safta, Dan Bolintineau, Craig M. Hamel, Gianluca Geraci, Sharlotte LB Kramer

Organizations: Sandia National Laboratories Livermore CA, USA · Sandia National Laboratories Albuquerque NM, USA

Abstract

Multifidelity modeling offers a practical route to accurate prediction and uncertainty quantification when high-fidelity data are scarce but lower-fidelity data are comparatively abundant. This work develops a bifidelity Langevin sampling hypernetwork framework that constructs probabilistic neu- ral response models by learning a posterior distribution over model parameters rather than a single deterministic fit. The method combines a flexible shared-latent data representation for low- and high-fidelity responses, enabling parameter sharing and transfer learning between the fidelities. We introduce several methodological enhancements to improve the robustness and efficiency of this ap- proach, including stability-promoting drift models, staged and conditional training strategies, adap- tive initialization of Langevin trajectories, and training procedures designed to better exploit corre- lations between fidelity levels. These additions improve convergence behavior, reduce the sampling effort required to obtain representative posterior ensembles, and provide more reliable predictive uncertainty in data-limited regimes. The resulting framework makes more effective use of available multifidelity data while preserving interpretability through the learned posterior structure, yielding improved predictive performance, better uncertainty characterization, and a more stable workflow for constructing uncertainty-aware surrogate models. The method is demonstrated on experimental, high-fidelity simulation and illustrative synthetic data.

Figures & tables

Explore similar work

CardsList
  1. Bifidelity Parameter Estimation Using Conditional Diffusion Models

    Apr 2, 2025Caroline Tatsuoka, Minglei Yang, Dongbin Xiu +1Simulation-Based InferenceSurrogate Modeling

  2. Multi-Fidelity Training Reshapes Uncertainty Decomposition in Atomistic Models

    Oct 26, 2026Matteo Tolloso, Davide Bacciu, Luigi BonatiMultifidelity ModelingUncertainty Calibration

  3. A Unified Hierarchical Multi-Task Multi-Fidelity Framework for Data-Efficient Surrogate Modeling in Manufacturing

    Mar 10, 2026Manan Mehta, Zhiqiao Dong, Yuhang Yang +1Surrogate ModelingMultifidelity Modeling