3–5 Jun 2026
Pisa
Europe/Rome timezone

Leveraging neural active manifolds for stratified sampling

4 Jun 2026, 17:30
15m
Pisa

Pisa

MS02 - Advances in Neural Network Approximation and Surrogate Modeling for Scientific Machine Learning MS02.2 - Advances in Neural Network Approximation and Surrogate Modeling for Scientific Machine Learning

Speaker

Andrea Zanoni (Scuola Normale Superiore)

Description

Propagating uncertainty from a potentially large number of random inputs through a computational model is becoming increasingly challenging due to the high cost of evaluating complex simulations. Stratified sampling is a well-known variance reduction strategy that, however, has mainly been employed in low-dimensional applications because of the difficulty of extending it to high-dimensional settings. In this talk, we propose using a recently introduced nonlinear dimensionality reduction approach, neural active manifolds (NeurAM), to enable stratified sampling in high dimensions. We leverage autoencoders to discover a one-dimensional manifold that captures most of the variability of the model output, aided by a simultaneously learned surrogate model whose inputs lie on this manifold. We then use the discovered neural active manifold to project a one-dimensional stratification back into the original input space, generating partitions that tend to follow the level sets of the model.

Authors

Andrea Zanoni (Scuola Normale Superiore) Daniele E. Schiavazzi (University of Notre Dame) Gianluca Geraci (Sandia National Laboratories)

Presentation materials

There are no materials yet.