Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “normalizing flows”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Variance reduction in lattice QCD observables via normalizing flows

Normalizing flows can be used to construct unbiased, reduced-variance estimators for lattice field theory observables that are defined by a derivative with respect to action parameters. This work implements the approach for observables involving gluonic operator insertions in the SU(3) Yang-Mills theory and two-flavor QCD in four space-time dimensions. Variance reduction by factors of 10–60 is achieved in glueball correlation functions and in gluonic matrix elements related to hadron structure, with demonstrated computational advantages. The observed variance reduction is found to be approximately independent of the lattice volume, so volume transfer can be utilized to minimize training costs.

Abbott, Ryan [Columbia U.; MIT, Cambridge, CTP; IA↗
DOE OSTI2025

Progress in Normalizing Flows for 4d Gauge Theories

Normalizing flows have arisen as a tool to accelerate Monte Carlo sampling for lattice field theories. This work reviews recent progress in applying normalizing flows to 4-dimensional nonabelian gauge theories, focusing on two advancements: an architectural improvement referred to as learned active loops, and the application of correlated ensemble methods to QCD with N f = 2 dynamical fermions.

Abbott, Ryan [Massachusetts Institute of Technolog↗
DOE OSTI2026

Monte Carlo Event Generation with Continuous Normalizing Flows

We apply continuous normalizing flows trained with the flow matching method to the problem of phase-space sampling in Monte Carlo event generation for high-energy collider physics. Focusing on lepton-pair and top-quark pair production with multiple jets, the two computationally most expensive processes at the Large Hadron Collider, we train helicity-conditioned continuous normalizing flows to remap the random numbers used in matrix element evaluation. Compared to standard methods, we achieve unweighting efficiency improvements by factors of up to 184 and 25 for the two processes at their respective highest jet number, at the cost of an increased evaluation time. When combining the advantages of continuous normalizing flows with the fast evaluation times of coupling-layer-based flows, using the RegFlow approach, we find parton-level unweighted event generation walltime gains of about a factor of 10 at the highest jet numbers. These substantial gains highlight the promise of samplers based on machine learning for next-generation collider experiments.

Bothmann, Enrico [CERN; Gottingen U.] (ORCID:00000↗
DOE OSTI2025

Conditional Pseudo-Reversible Normalizing Flow for Surrogate Modeling in Quantifying Uncertainty Propagation

We introduce a conditional pseudo-reversible normalizing flow (PR-NF) that directly learns conditional probability distributions from noisy physical models to efficiently quantify both forward and inverse uncertainty propagation. Traditional surrogate modeling approaches approximate only the deterministic component of physical models, requiring separate noise characterization and computationally expensive sampling methods for inverse problems. Here, in this work, we develop the conditional PR-NF model to directly learn and efficiently generate samples from the conditional probability density functions (PDFs). The training process utilizes dataset consisting of input-output pairs without requiring prior knowledge about the noise and the function. Once trained, our model efficiently generates samples from conditional PDFs for any input within the training domain. Moreover, the pseudo-reversibility feature allows for the use of fully connected neural network architectures, which simplifies the implementation and enables theoretical analysis. We provide a rigorous convergence analysis of the conditional PR-NF model, showing its ability to converge to the target conditional PDF using the Kullback−Leibler divergence. To demonstrate the effectiveness of our method, we apply it to several benchmark tests and a real-world geologic carbon storage problem.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Normalizing flows for high-dimensional detector simulations

Whenever invertible generative networks are needed for LHC physics, normalizing flows show excellent performance. In this work, we investigate their performance for fast calorimeter shower simulations with increasing phase space dimension. We use fast and expressive coupling spline transformations applied to the CaloChallenge datasets. In addition to the base flow architecture we also employ a VAE to compress the dimensionality and train a generative network in the latent space. We evaluate our networks on several metrics, including high-level features, classifiers, and generation timing. Our findings demonstrate that invertible neural networks have competitive performance when compared to autoregressive flows, while being substantially faster during generation.

Ernst, Florian↗
DOE OSTI2026

Unbinned extraction of $γ$ from $B\to DK$ with normalizing flows

We introduce an unbinned method for extracting the CKM angle $γ$ from the decay chain $B^\pm \to (D \to K_S π^+ π^-) K^\pm$ using normalizing flows (NFs). The NFs, trained on $D$ decay data, learn a faithful continuous representation of the amplitude and strong phase variation over the $D\to K_Sπ^+π^-$ Dalitz plot whose fidelity improves with increased data sample sizes. With this input, the $B$ decay data can be used to extract the parameters $r_B$, $δ_B$, and $γ$. We test the method on Monte Carlo generated data, where it successfully recovers the injected value of $γ$ within uncertainties. The present implementation propagates statistical uncertainties from finite training data via an ensemble of independently trained flows, and does not attempt to capture the effects of systematic experimental errors. We explore two versions of the method that differ in how the trigonometric constraint on phase variation is encoded, and comment on the possible extension to Bayesian NFs, which would provide direct uncertainty estimates on the learned densities without requiring ensemble training.

Grossman, Yuval [Cornell U., LEPP]↗
DOE OSTI2025

Unifying simulation and inference with normalizing flows

There have been many applications of deep neural networks to detector calibrations and a growing number of studies that propose deep generative models as automated fast detector simulators. We show that these two tasks can be unified by using maximum likelihood estimation (MLE) from conditional generative models for energy regression. Unlike direct regression techniques, the MLE approach is prior independent and non-Gaussian resolutions can be determined from the shape of the likelihood near the maximum. Using an ATLAS-like calorimeter simulation, we demonstrate this concept in the context of calorimeter energy calibration. Published by the American Physical Society 2025

Hadronic calorimiters↗
DOE OSTI2025

Solving high-dimensional inverse problems using amortized likelihood-free inference with noisy and incomplete data

Here, we present a likelihood-free probabilistic inversion method based on normalizing flows for high-dimensional inverse problems. The proposed method is composed of two complementary networks: a summary network for data compression and an inference network for parameter estimation. The summary network encodes raw observations into a fixed-size vector of summary features, while the inference network generates samples of the approximate posterior distribution of the model parameters based on these summary features. The posterior samples are produced in a deep generative fashion by sampling from a latent Gaussian distribution and passing these samples through an invertible transformation. We construct this invertible transformation by sequentially alternating conditional invertible neural network and conditional neural spline flow layers. The summary and inference networks are trained simultaneously. We apply the proposed method to an inversion problem in groundwater hydrology to estimate the posterior distribution of the log-conductivity field conditioned on spatially sparse time-series observations of the system’s hydraulic head responses. The conductivity field is represented with 706 degrees of freedom in the considered problem. Comparison with the likelihood-based iterative ensemble smoother PEST-IES method demonstrates that the proposed method accurately estimates the parameter posterior distribution and the observations’ predictive posterior distribution at a fraction of the inference time of PEST-IES.

conditional invertible neural network↗
DOE OSTI2026

Neural posterior unfolding

Differential cross section measurements are the currency of scientific exchange in particle and nuclear physics. A key challenge for these analyses is the correction for detector distortions, known as deconvolution or unfolding. Binned unfolding of cross section measurements traditionally rely on the regularized inversion of the response matrix that represents the detector response, mapping pre-detector (`particle level') observables to post-detector (`detector level') observables. In this paper we introduce Neural Posterior Unfolding, a modern, Bayesian approach that leverages normalizing flows for unfolding. By using normalizing flows for neural posterior estimation, NPU offers several key advantages including implicit regularization through the neural network architecture, fast amortized inference that eliminates the need for repeated retraining, and direct access to the full uncertainty in the unfolded result. In addition to introducing NPU, we implement a classical Bayesian unfolding method called Fully Bayesian Unfolding (FBU) in modern Python so it can also be studied. These tools are validated on simple Gaussian examples and then tested on simulated jet substructure examples from the Large Hadron Collider (LHC). We find that the Bayesian methods are effective and worth additional development to be analysis ready for cross section measurements at the LHC and beyond.

Analysis and statistical methods↗
DOE OSTI2026

Efficient many-jet event generation with flow matching

We apply for the first time, to the best of our knowledge, the flow matching method to the problem of phase-space sampling for event generation in high-energy collider physics. By training the model to remap the random numbers used to generate the momenta and helicities of the scattering matrix elements as implemented in the portable partonic event generator pepper, we find substantial efficiency improvements in the studied processes. We focus our study on the highest final-state multiplicities in Drell-Yan and top-antitop pair production used in simulated samples for the Large Hadron Collider, which computationally are the most relevant ones. We find that the unweighting efficiencies improve by factors of 184 and 25, respectively, when compared to the standard approach of using a vegas-based optimization. We also compare continuous normalizing flows trained with flow matching against the previously studied normalizing flows based on coupling layers and find that the former leads to better results, faster training and a better scaling behavior across the studied multiplicity range, while the latter evaluate faster. When combining the advantages of both methods using the regflow approach, we find parton-level unweighted event generation walltime gains of about a factor of 10 at the highest final-state multiplicities.

Bothmann, E. [CERN; Gottingen U.] (ORCID:000000016↗
DOE OSTI2025

Flow annealed importance sampling bootstrap meets differentiable particle physics

High-energy physics requires the generation of large numbers of simulated data samples from complex but analytically tractable distributions called matrix elements. Surrogate models, such as normalizing flows, are gaining popularity for this task due to their computational efficiency. We adopt an approach based on flow annealed importance sampling bootstrap (FAB) that evaluates the differentiable target density during training and helps avoid the costly generation of training data in advance. We show that FAB reaches higher sampling efficiency with fewer target evaluations in high dimensions in comparison to other methods.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2026

Conditional deep generative models for simultaneous simulation and reconstruction of entire events

We extend the particle-flow neural assisted simulations (arnassus) framework of fast simulation and reconstruction to entire collider events. In particular, we use two generative artificial intelligence tools, continuous normalizing flows and diffusion models, to create a set of reconstructed particle-flow objects conditioned on truth-level particles from CMS Open Simulations. While previous work focused on jets, our updated methods now can accommodate all particle-flow objects in an event along with particle-level attributes like particle type and production vertex coordinates. This approach is fully automated, entirely written in Python, and GPU-compatible. Using a variety of physics processes at the LHC, we show that the extended arnassus is able to generalize beyond the training dataset and outperforms the standard, public tool elphes.

Dreyer, Etienne [Weizmann Institute of Science, Re↗
DOE OSTI2025

Flow matching beyond kinematics: Generating jets with particle identification and trajectory displacement information

We introduce the first generative model trained on the etlass dataset. Our model generates jets at the constituent level, and it is a permutation-equivariant continuous normalizing flow (CNF) trained with the flow matching technique. It is conditioned on the jet type, so that a single model can be used to generate the ten different jet types of etlass. For the first time, we also introduce a generative model that goes beyond the kinematic features of jet constituents. The etlass dataset includes more features, such as particle-ID and track impact parameter, and we demonstrate that our CNF can accurately model all of these additional features as well. Our generative model for etlass expands on the versatility of existing jet generation techniques, enhancing their potential utility in high-energy physics research, and offering a more comprehensive understanding of the generated jets. Published by the American Physical Society 2025

Birk, Joschka (ORCID:0000000219310127)↗
DOE OSTI2025

Reweighting configurations generated by transferable, machine learned models for protein sidechain backmapping

Multiscale modeling requires the linking of models at different levels of detail, with the goal of gaining accelerations from lower fidelity models while recovering fine details from higher resolution models. Communication across resolutions is particularly important in modeling soft matter, where tight couplings exist between molecular-level details and mesoscale structures. While multiscale modeling of biomolecules has become a critical component in exploring their structure and self-assembly, backmapping from coarse-grained to fine-grained, or atomistic, representations presents a challenge, despite recent advances through machine learning. A major hurdle, especially for strategies utilizing machine learning, is that backmappings can only approximately recover the atomistic ensemble of interest. We demonstrate conditions for which backmapped configurations may be reweighted to exactly recover the desired atomistic ensemble. By training separate decoding models for each sidechain type, we develop an algorithm based on normalizing flows and geometric algebra attention to autoregressively propose backmapped configurations for any protein sequence. Critical for reweighting with modern protein force fields, our trained models include all hydrogen atoms in the backmapping and make probabilities associated with atomistic configurations directly accessible. We also demonstrate, however, that reweighting is extremely challenging despite state-of-the-art performance on recently developed metrics and generation of configurations with low energies in atomistic protein force fields. Through detailed analysis of configurational weights, we show that machine-learned backmappings must not only generate configurations with reasonable energies, but also correctly assign relative probabilities under the generative model. These are broadly important considerations in generative modeling of atomistic molecular configurations.

Monroe, Jacob I. [Univ. of Arkansas, Fayetteville,↗
DOE OSTI2025

ForestFlow: predicting the Lyman-α forest clustering from linear to nonlinear scales

On large scales, the Lyman-α forest provides insights into the expansion history of the Universe, while on small scales, it imposes strict constraints on the growth history, the nature of dark matter, and the sum of neutrino masses. This work introduces ForestFlow, a novel framework that bridges the gap between large- and small-scale analyses, which have traditionally relied on distinct modeling approaches. Using conditional normalizing flows, ForestFlow predicts the two Lyman-α linear biases (b δ and b η ) and six parameters describing small-scale deviations of the three-dimensional flux power spectrum (P 3D ) from linear theory as a function of cosmology and intergalactic medium physics. These are then combined with a Boltzmann solver to make consistent predictions, from arbitrarily large scales down to the nonlinear regime, for P 3D and any other statistics derived from it. Trained on a suite of 30 fixed-and-paired cosmological hydrodynamical simulations spanning redshifts from z = 2 to 4.5, ForestFlow achieves 3 and 1.5% precision in describing P 3D and the one-dimensional flux power spectrum (P 1D ) from linear scales to k = 5 Mpc −1 and k ∥ = 4 Mpc −1 , respectively. Thanks to its conditional parameterization, ForestFlow shows similar performance for ionization histories and two ΛCDM model extensions – massive neutrinos and curvature – even though none of these are included in the training set. This framework will enable full-scale cosmological analyses of Lyman-α forest measurements from the DESI survey.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

CaloChallenge 2022: a community challenge for fast calorimeter simulation

Here, we present the results of the ‘Fast Calorimeter Simulation Challenge 2022’—the CaloChallenge. We study state-of-the-art generative models on four calorimeter shower datasets of increasing dimensionality, ranging from a few hundred voxels to a few tens of thousand voxels. The 31 individual submissions span a wide range of current popular generative architectures, including variational autoencoders (VAEs), generative adversarial networks (GANs), normalizing flows, diffusion models, and models based on conditional flow matching. We compare all submissions in terms of quality of generated calorimeter showers, as well as shower generation time and model size. To assess the quality we use a broad range of different metrics including differences in one-dimensional histograms of observables, KPD/FPD scores, AUCs of binary classifiers, and the log-posterior of a multiclass classifier. The results of the CaloChallenge provide the most complete and comprehensive survey of cutting-edge approaches to calorimeter fast simulation to date. In addition, our work provides a uniquely detailed perspective on the important problem of how to evaluate generative models. As such, the results presented here should be applicable for other domains that use generative AI and require fast and faithful generation of samples in a large phase space.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2026

Dark Energy Survey Year 3 results: Simulation-based 𝑤CDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design

Data-driven approaches using deep learning are emerging as powerful techniques to extract non-Gaussian information from cosmological large-scale structure. Here, this work presents the first simulation-based inference (SBI) pipeline that combines weak lensing and galaxy clustering maps in a realistic Dark Energy Survey Year 3 (DES Y3) configuration and serves as preparation for a forthcoming analysis of the survey data. We develop a scalable forward model based on the CosmoGridV1 suite of N-body simulations to generate over one million self-consistent mock realizations of DES Y3 at the map level. Leveraging this large dataset, we train deep graph convolutional neural networks on the full survey footprint in spherical geometry to learn low-dimensional features that approximately maximize mutual information with target parameters. These learned compressions enable neural density estimation of the implicit likelihood via normalizing flows in a ten-dimensional parameter space spanning cosmological 𝑤CDM, intrinsic alignment, and linear galaxy bias parameters, while marginalizing over baryonic, photometric redshift, and shear bias nuisances. To ensure robustness, we extensively validate our inference pipeline using synthetic observations derived from both systematic contaminations in our forward model and independent Buzzard galaxy catalogs. Our forecasts yield significant improvements in cosmological parameter constraints, achieving 2−3× higher figures of merit in the 𝛺 𝑚 − 𝑆 8 plane relative to our implementation of baseline two-point statistics and effectively breaking parameter degeneracies through probe combination. These results demonstrate the potential of SBI analyses powered by deep learning for upcoming Stage-IV wide-field imaging surveys.

Thomsen, A. [Zurich, ETH] (ORCID:0000000203099021)↗
DOE OSTI2025

Efficient Monte Carlo event generation for neutrino-nucleus exclusive cross sections

Modern neutrino-nucleus cross section computations need to incorporate sophisticated nuclear models to achieve greater predictive precision. However, the computational complexity of these advanced models often limits their practicality for experimental analyses. To address this challenge, we introduce a new Monte Carlo method utilizing normalizing flows to generate surrogate cross sections that closely approximate those of the original model while significantly reducing computational overhead. As a case study, we built a Monte Carlo event generator for the neutrino-nucleus cross section model developed by the Ghent group. This model employs a Hartree-Fock procedure to establish a quantum mechanical framework in which both the bound and scattering nucleon states are solutions to the mean-field nuclear potential. The surrogate cross sections generated by our method demonstrate excellent accuracy with a relative effective sample size of more than 98.4%, providing a computationally efficient alternative to traditional Monte Carlo sampling methods for differential cross sections.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗