Factorizable Normalizing Flows for parameter-dependent density morphing
Davide Valsecchi, Mauro Donegà, Rainer Wallny
stat.ML, cs.LG, hep-ex, hep-th, physics.data-an
Submitted: 2026-06-29
Comments: 14 pages, 8 figures. Code: https://doi.org/10.5281/zenodo.21011625
License: http://creativecommons.org/licenses/by/4.0/
The gist: Normalizing Flows excel at modeling a single fixed density, yet many problems across the sciences, such as high energy physics, instead require modeling how that density deforms as a function of
Terminology
Abstract
Normalizing Flows excel at modeling a single fixed density, yet many problems across the sciences, such as high energy physics, instead require modeling how that density deforms as a function of continuous parameters: the strength of a physical effect, a calibration constant, or a source of systematic uncertainty. Learning a separate flow for every parameter configuration quickly becomes intractable, since the number of joint settings grows exponentially with the number of parameters. We introduce Factorizable Normalizing Flows (FNFs), which represent the parameter-dependent density as a fixed, high-fidelity flow for a reference configuration composed with a learnable transformation that is polynomial in the parameters and factorized over them. This structure has a practical consequence: each parameter's effect is learned in isolation, from samples in which that parameter alone is varied. The combined response of many parameters is then recovered by summation at inference, without ever sampling their combinatorially large joint space. On a controlled problem with two interpretable deformations applied jointly to the data, the learned transformation reproduces the true deformations and matches the optimal likelihood, while optional interaction terms capture residual correlations when several parameters vary strongly at once. The resulting model is interpretable, scales linearly with the number of parameters, and keeps the likelihood tractable. This provides a general tool for any inference workflow requiring continuous density morphing, and directly enables the next generation of unbinned likelihood fits in high energy physics.
Sources
- Masked Autoregressive Flow for Density Estimation
- Neural Spline Flows
- Decoupled Weight Decay Regularization
- Convex Potential Flows: Universal Probability Distributions with Optimal Transport and Convex Optimization
- Input Convex Neural Networks
- Analysis-ready Generative Unfolding
- Simulation-Prior Independent Neural Unfolding Procedure
Related papers
- Behavior of prediction performance metrics with rare events
- Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
- A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
- Scalable Posterior Uncertainty for Flexible Density-Based Clustering
- A Transport Geometry of Benign Overfitting
- Efficient quadratic entropy with distance sketches