arXiv AI By Exauc\'e Luweh Adjim Ngarti (AIRSEA), \'Elise Arnaud (AIRSEA), L\'eo Nicoletti (Mines Saint-\'Etienne MSE, LIMOS, FAYOL-ENSMSE, FAYOL-ENSMSE), Victor Trappler (Mines Saint-\'Etienne MSE, LIMOS, FAYOL-ENSMSE, FAYOL-ENSMSE), Arthur Vidard (AIRSEA)

Amortized Localized Posterior Inference for Model Calibration with Application to an Ocean Turbulence Closure

Read the original on arXiv AI →

The paper introduces a method for localized posterior inference that weights observation likelihoods based on proximity to a target context, using Gaussian localization and a conditional normalizing flow for amortized variational inference. By integrating these localized posteriors over a context measure, the authors obtain a marginalized parameter distribution that accounts for observation and context uncertainties. Experiments with the Branin-Hoo model and a k‑ε turbulence closure in an ocean model demonstrate that localization reduces aggregate predictive errors and improves temperature and zonal‑velocity predictions compared to global or deterministic calibration.

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 AI.

arXiv AI
Oct 28

R\`enyi-Regularized Flow-Based Variational Inference for Bayesian Calibration

arXiv:2610. 37169v1 Announce Type: new Abstract: Variational inference with normalizing flows provides a flexible posterior approximation for Bayesian model calibration, but struggles when the posterior has several modes separated by regions of low probability mass.

By Exauc\`e Luweh Adjim Ngarti (AIRSEA), \'Elise Arnaud (AIRSEA), L\'eo Nicoletti (Mines Saint-\'Etienne MSE, LIMOS, FAYOL-ENSMSE, FAYOL-ENSMSE), Victor Trappler (Mines Saint-\'Etienne MSE, LIMOS, FAYOL-ENSMSE, FAYOL-ENSMSE), Arthur Vidard (AIRSEA)
Hugging Face Trending Papers
Oct 27

Rènyi-Regularized Flow-Based Variational Inference for Bayesian Calibration

The paper investigates how variational inference using normalizing flows can struggle with multimodal posteriors, particularly when modes are separated by low-probability regions. By adding a Rényi divergence-based regularization term, the authors encourage the approximate posterior to cover the prior support, reducing mode collapse from 50% to 25% in a twin experiment. However, this regularization alters the Bayesian objective and introduces a trade‑off in approximation accuracy, and it does not solve mode collapse in general.

arXiv Machine Learning
Nov 5

FlowReconBench: Accuracy, calibration and cost of learned and classical flow reconstruction from sparse sensors

FlowReconBench compares ten learned and five classical flow reconstruction methods across three flow regimes—laminar cylinder wake, 2D Kolmogorov turbulence, and 3D isotropic turbulence—using consistent data splits and sensor draws. The benchmark evaluates per‑channel error, ensemble spread, calibration, spectra, and computational cost, revealing that classical estimators excel in low‑order and dense‑sensing scenarios, while data‑driven models outperform when sensor density or observability decreases. Generative models provide broader but underdispersed ensembles, and their performance can be tuned via sampler budget, though at higher computational expense.

By Nicolas Tricard, Linzheng Wang, Jason Chen, Sili Deng