arXiv Statistics ML By Owen Davis, Daniel Sharp, Youssef Marzouk, Gianluca Geraci

Multifidelity Formulations for Triangular Transport

Read the original on arXiv Statistics ML →

The paper introduces multifidelity techniques for building triangular transport maps when high‑fidelity data are limited but low‑fidelity data are plentiful. Two strategies are proposed: a hierarchical approach that composes maps across fidelity levels, and a non‑hierarchical method that uses monotonicity‑preserving corrections to incorporate low‑fidelity information. Numerical tests show these methods outperform single‑fidelity transport and, when applied to amortized simulation‑based inference, improve conditional sampling and uncertainty quantification in data‑scarce regimes.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Statistics ML.

Hugging Face Trending Papers
Jun 29

The Fundamental Limits of Valid Transport Map Estimation

Many modern generative modeling methods, including diffusion models, normalizing flows, and flow matching, estimate transport maps or plans between distributions without explicitly targeting an optimal transport (OT) map. In applications like generative modeling, the transport cost itself is irrelevant, and this makes it natural to target maps which are more tractable from either a statistical or computational standpoint.