Search NASA⌕ Search

SEARCH · Search NASA

Results for “generative machine learning”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

MULTI-LEADER: MULTI-source LEarning-Accelerated Design of high-Efficiency multi-stage compRessor (Final Technical Report)

The objective of MULTI-LEADER is to cut design costs by 80% while generating more energy-efficient designs of multi-stage compressors by developing and implementing novel machine learning (ML) techniques, which enable faster and fewer design iterations, improved solver performance, and concurrent multi-disciplinary design. Current industrial practices for the design of multi-stage compressors involve simulation-based design optimization with successive levels of model fidelity, iteratively evaluated between distinct disciplines, one stage at a time to tackle the high dimensional design variations. This project addresses these key design challenges: (1) concurrent optimization of multiple stages under many non-linear constraints; (2) multitude of evaluation of high-fidelity and expensive solvers and their gradients during optimization convergence in high-dimensional design; (3) multi-disciplinary design to maximize aerodynamic performance while guaranteeing structural integrity and additive manufacturability; (4) utilization of multiple fidelity of solvers with disparate parameterization and modeling assumptions. MULTI-LEADER achieved more than 5x speed up in detailed design of more energy-efficient compressors via these machine learning (ML) innovations: (i) rapid design surrogates by multi-source learning from diverse fidelities across multiple disciplines, (ii) physics-constrained data-augmented modeling for improved empiricism, (iii) generative manifold embedding for high dimensional concurrent design without gradient information; (iv) budget-constrained fidelity-adaptive sampling towards fewer design iterations.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2025

Convergent Protocols for Computing Protein–Ligand Interaction Energies Using Fragment-Based Quantum Chemistry

Fragment-based quantum chemistry methods offer a way to sidestep the steep nonlinear scaling of electronic structure calculations so that large molecular systems can be investigated using high-level methods. Here, we use fragmentation to compute protein–ligand interaction energies in systems with several thousand atoms, using a new software platform for managing fragment-based calculations that implements a screened many-body expansion. Convergence tests using a minimal-basis semiempirical method (HF-3c) indicate that two-body calculations, with single-residue fragments and simple hydrogen caps, are sufficient to reproduce interaction energies obtained using conventional supramolecular electronic structure calculations, to within 1 kcal/mol at about 1% of the computational cost. We also demonstrate that the HF-3c results are illustrative of trends obtained with density functional theory in basis sets up to augmented quadruple-ζ quality. Strategic deployment of fragmentation facilitates the use of converged biomolecular model systems alongside high-quality electronic structure methods and basis sets, bringing ab initio quantum chemistry to systems of hitherto unimaginable size. This will be useful for generation of high-quality training data for machine learning applications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

The Bias-Variance-Correlation Tradeoff and Its Implications for ML Applications in HEP

The bias-variance tradeoff is a well-recognized phenomenon in statistics and machine learning. In this talk, I will discuss an extension, dubbed the bias-variance-correlation tradeoff. Roughly speaking, as the flexibility of a model decreases, the correlations in the outputs of a trained model for different inputs increases. Such correlations have implications for several applications of machine learning in high energy physics, e.g., the use generative models for event generation. In particular, I will argue that claims in the literature of data amplification by generative models stem from ignoring important correlations between the model's outputs for different inputs.

Shyamsundar, Prasanth [Fermilab] (ORCID:0000000227↗
DOE OSTI2026

Assessing shellfish water exposure to fecal bacteria pollution in Salish Sea: three-dimensional modeling and implications for monitoring

Fecal bacteria (FB) contamination poses significant risks to shellfish safety and management in coastal and estuarine waters. Despite extensive pollution identification and correction efforts, FB contamination in shellfish-growing areas persists in the Salish Sea, highlighting the need to identify overlooked sources and better understand FB transport from riverine and shoreline inputs to shellfish beds. To address this, a high-resolution three-dimensional hydrodynamic model coupled with FB kinetics was developed and applied to a case study site in Salish Sea—Portage Bay—to simulate freshwater plume circulation, flushing dynamics, and bacterial transport. Daily FB loading from the major freshwater inflow—Nooksack River was generated by both linear interpolation and integrating a machine learning approach (XGBoost), trained on historical hydrological and meteorological data. The model successfully reproduced both the magnitude and seasonal variation of FB concentrations in Portage Bay for the year of 2021, demonstrating that simplified FB kinetics with first-order decay due to mortality was effective in this dynamic coastal environment with short flushing time. Model results identified the Nooksack River as the dominant far-field FB source, while scenario simulations showed that near-field coastal stormwater outfalls elevated local FB levels following rainfall, particularly under low-flow conditions. The XGBoost prediction provided comparable or superior accuracy to linear interpolation, particularly during periods of missing observational data, by capturing short-term variability and event-driven loading more effectively. Integrating data-driven riverine FB inputs with mechanistic coastal numerical modeling provides a robust framework for operational forecasting of shellfish bed exposure risk and supports adaptive monitoring and management of shellfish growing areas in the Salish Sea and similar coastal systems.

Salish Sea↗
DOE OSTI2025

Sub-microsecond Transformers for Jet Tagging on FPGAs

We present the first sub-microsecond transformer implementation on an FPGA achieving competitive performance for state-of-the-art high-energy physics benchmarks. Transformers have shown exceptional performance on multiple tasks in modern machine learning applications, including jet tagging at the CERN Large Hadron Collider (LHC). However, their computational complexity prohibits use in real-time applications, such as the hardware trigger system of the collider experiments up until now. In this work, we demonstrate the first application of transformers for jet tagging on FPGAs, achieving $\mathcal{O}(100)$ nanosecond latency with superior performance compared to alternative baseline models. We leverage high-granularity quantization and distributed arithmetic optimization to fit the entire transformer model on a single FPGA, achieving the required throughput and latency. Furthermore, we add multi-head attention and linear attention support to hls4ml, making our work accessible to the broader fast machine learning community. This work advances the next-generation trigger systems for the High Luminosity LHC, enabling the use of transformers for real-time applications in high-energy physics and beyond.

Laatu, Lauri [Imperial Coll., London]↗
DOE OSTI2025

Automated Nanocrystal Synthesis: Lessons from 25 Years of Robots, Microfluidics, and Machine Learning

Here, this perspective highlights the evolution of techniques for automating the synthesis of colloidal nanocrystals. Over the past 25 years, microfluidic reactors and robotic workflows have been developed to enhance the reproducibility of nanocrystal synthesis, facilitate rapid screening of reaction conditions, optimize material properties, and perform multistep syntheses of high-quality nanoparticles with complex heterostructures. Modern automated systems are now valued for their ability to generate robust data sets for validating physical models, supporting chemical mechanisms, training machine learning models, and for directing autonomous experimentation. We discuss the early challenges and limitations of these technologies and present key lessons for effectively utilizing automated and ML-guided tools to accelerate nanocrystal discovery for the next 25 years.

Nanocrystals↗
DOE OSTI2026

EXERGETIC: De-Risking Next-Generation Resilient Geothermal Hybrids via At-Scale Evaluation Using Virtual Emulation Digital Twin Environment for Efficient Operation

The DOE-GTO-funded project, award number 5.1.2.12, entitled "EXERGETIC - De-risking Next Generation Resilient Geothermal Hybrids via at-Scale Evaluation Using a Virtual Emulation Digital Twin Environment for Efficient Operation," advances the solution to these challenges by developing and validating a geothermal co-emulation environment implemented at the National Laboratory of the Rockies (NLR)'s Advanced Research on Integrated Energy Systems (ARIES) platform. This framework enables the de-risking of next-generation geothermal and geothermal hybrid systems through high-fidelity modeling, real-time digital emulation, advanced control strategies, and techno-economic assessment. The project focused on geothermal hybrid configurations that integrate geothermal power plants with concentrated solar power and underground thermal energy storage, enabling enhanced efficiency, flexibility, and grid support capabilities. The main goal of this project was the development of a geothermal digital co-emulation environment to demonstrate the technical and economic value of geothermal hybrid systems and their contribution to grid stability and flexibility. The EXERGETIC framework combined physics-based models, controls, and real assets at ARIES, including digital real-time simulators (DRTS), a 20-MW-scale controllable grid interface (CGI), and a 2-MW conventional generator. Detailed transient models were developed for the key subsystems of a hybrid geothermal plant, including parabolic trough solar collectors, reservoir thermal energy storage (RTES), and a binary Organic Rankine Cycle (ORC) power plant. The ORC model explicitly captured thermal inertia and off-design operation and integrated control strategies to dynamically respond to electric load profiles. The models were validated against published experimental and numerical studies, demonstrating strong agreement and confirming the accuracy and robustness of the modeling approach. The resulting digital twin represents geothermal-solar-storage systems at multiple scales (1 MW to 100 MW) and enables realistic emulation of grid-connected operation. The control architecture allows the geothermal resource to provide stable baseload generation, while solar and stored thermal energy supply flexible, dispatchable support during periods of high demand or variable grid conditions. A key contribution of the EXERGETIC project is the demonstration that geothermal hybrid systems can be designed to be active grid assets rather than passive baseload generators. Using the ARIES platform, the digital twin was evaluated under multiple grid scenarios, including load following, voltage support at the distribution level, and frequency response at the transmission level. Results show that hybrid geothermal systems can respond effectively to dynamic grid conditions, providing inertia-like behavior, primary frequency support, and voltage regulation through coordinated control. In addition to the performance and grid services capability analysis of geothermal and hybrid geothermal systems, the EXERGETIC project also focused on scalability and techno-economic analysis of geothermal hybrid plants. In particular, for the scalability analysis, machine-learning (ML)-based surrogate models were trained using data generated from the geothermal digital twin under different grid-connected scenarios and plant capacities. These ML models demonstrated strong interpolation and extrapolation capabilities across plant sizes, accurately reproducing both steady-state and transient responses with very low errors. Regarding the techno-economic analysis, plant performance results were integrated with cost models for hybrid geothermal systems, and the levelized cost of electricity (LCOE) was used as the main economic metric to evaluate system performance across a range of system capacities, solar shares, solar multiples, and storage durations. Results indicate that economies of scale significantly reduce geothermal LCOE as plant capacity increases, with large-scale systems (25-100 MW) achieving substantially lower costs than small plants. Hybridization with solar thermal energy and storage further improves economic performance by increasing capacity utilization and enabling flexible dispatch. In addition, thermal storage plays a critical role in reducing LCOE by maximizing geothermal, solar, and stored energy resources. In summary, the results from this project demonstrate that geothermal hybrid systems represent a promising alternative for increasing the energy conversion efficiency of geothermal technologies, contributing to the preservation of geothermal resources, and supporting the transition of geothermal plants from traditional baseload resources into flexible, resilient, and cost-competitive energy conversion technologies.

15 GEOTHERMAL ENERGY↗
DOE OSTI2025

Scalability analysis of heavy-duty gas turbines using data-driven machine learning

With the increasing integration of variable renewable energy sources into power systems, the role of flexible power generation technologies like gas turbines (GT) in rapid grid balancing remains crucial. This sustained importance underscores the need for scaled and precise modeling of GT to ensure effective integration within evolving energy frameworks. While physics-driven GT models integrate thermodynamics, fluid dynamics, and combustion principles, they often rely on approximate mathematical representations to accommodate scaling that may not capture the actual complex dynamics for GTs and inertial effects associated to GTs with different ratings. In this study, a data-driven model is proposed using machine learning (ML) techniques to conduct GT scalability analysis and performance evaluation with high accuracy. The ML model, trained on data from various operating conditions and performance parameters, aims to uncover intricate relationships and patterns, resembling GT characteristics at different scales (ratings). The model is developed to capture complex system interaction and to adapt to changing operational scenarios at different capacities, providing valuable insights of power system dynamics. In this study, the real-time digital simulator platform was employed to generate training data for the ML model and assess its dynamic characteristics. The ultimate objective was to develop a detailed modeling framework based on governing equations and data-driven ML capable of predicting key performance indicators, in thermal systems such as GTs, including power output, speed, fuel consumption, and exhaust temperature under diverse operating conditions at different scales. The developed ML framework demonstrated high accuracy, with mean relative errors for GT power prediction, reference speed, exhaust temperature, and compressor pressure ratio (CPR) parameters consistently below 0.1% across typical load fluctuation scenarios. Maximum deviations were limited to approximately 0.5 K for exhaust temperature and 0.009 for CPR, underscoring the model’s ability to replicating dynamic GT behavior with high precision. The adaptability of the ML model enables its application across diverse operational conditions and its extension to other thermal systems. By leveraging advanced ML techniques, this study presents a robust and scalable modeling framework that enhances GT simulation precision, facilitating improved integration into evolving power systems.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Probing degradation at solid-state battery interfaces using machine-learning interatomic potential

Solid-state batteries featuring fast ion-conducting solid electrolytes are promising next-generation energy storage technologies, yet challenges remain for practical deployment due to electro-chemo-mechanical instabilities at solid-solid interfaces. These interfaces, which include homogeneous/internal interfaces such as grain boundaries (GBs) and heterogeneous/external interfaces between solid-electrolyte and electrode materials, can impede Li-ion transport, deteriorate performance, and eventually lead to cell failure. Here, in this study, we leverage large-scale molecular simulations, enabled by validated machine-learning interatomic potentials, to directly probe the onset of interfacial degradation at the garnet Li 7 La 3 Zr 2 O 12 (LLZO) solid-electrolyte/LiCoO 2 (LCO) cathode interface. By surveying different interfacial geometries and compositions, it is found that Li-deficient interfaces can lead to severe interfacial disordering with cation mixing and Co interdiffusion from LCO into LLZO. By contrast, Li-sufficient interfaces are less disordered, although elemental segregation with local ordering is observed. As a consequence of Co interdiffusion, Co-rich regions are formed at the GBs of LLZO due to cation segregation and trapping effects. This behavior is independent of the GB tilting axis, degree of disorder at the GBs, and Co concentration, which implies Co clustering at GBs is a general phenomenon in polycrystalline LLZO and can dictate its overall transport and mechanical properties. Our findings elucidate the underlying fundamental mechanisms that give rise to experimentally observed physicochemical properties and provide guidelines for interface design that can mitigate interfacial degradation and improve cycling performance.

25 ENERGY STORAGE↗
DOE OSTI2025

Machine learning-enabled discovery of ionic liquid–solvent electrolytes exhibiting high ionic conductivity

Ionic liquids (ILs), which are a class of materials with versatile nature and growing popularity, are facing impediments toward widespread usage as electrolytes due to various factors such as low ionic conductivity, high viscosity, high market price etc. One of the ways these limitations can be addressed is by mixing ILs with a molecular solvent. In a combinatorial sense, there exists an immense number of specific IL–solvent combinations. An exhaustive experimental or even simulation-based investigation of the chemical space spanned by such combinations can be extremely time-consuming, expensive, and nearly impossible. An alternative approach is to employ machine learning-based models developed from available databases. Although there exists prior literature that integrates machine learning to investigate mixtures of specific solvents with ILs, these models lack generalization necessitating development of a large number of ML models to handle various solvents. To remedy this shortcoming, as a part of designing green electrolytes with high ionic conductivity that can have potential applications in next-generation batteries and solar cells, this work aims to develop a unified machine learning model to predict ionic conductivity of any IL–solvent mixture system. In this regard, three models, namely, Random Forest, extreme gradient boosting (XGBoost), and artificial neural network (ANN) were formulated using the NIST ILThermo database. The dataset contained 549 unique ionic liquids from 16 cation families and 81 unique solvents, representing a total of 23 712 datapoints. SHAPLEY additive explanation (SHAP) method was used to assess the impact of various features on model prediction and their significance was compared with literature to gain physical insight about the model behavior. Finally, using the developed models, approximately 2.5 million IL–solvent mixtures at five different compositions were screened at room temperature. The high-throughput screening yielded nearly 19 000 IL–solvent mixtures for which ionic conductivity was found to exceed the ionic conductivity of conventional Li-ion battery electrolyte.

25 ENERGY STORAGE↗
DOE OSTI2025

Data imbalance in drug response prediction: multi-objective optimization approach in deep learning setting

Abstract Drug response prediction (DRP) methods tackle the complex task of associating the effectiveness of small molecules with the specific genetic makeup of the patient. Anti-cancer DRP is a particularly challenging task requiring costly experiments as underlying pathogenic mechanisms are broad and associated with multiple genomic pathways. The scientific community has exerted significant efforts to generate public drug screening datasets, giving a path to various machine learning models that attempt to reason over complex data space of small compounds and biological characteristics of tumors. However, the data depth is still lacking compared to application domains like computer vision or natural language processing domains, limiting current learning capabilities. To combat this issue and improves the generalizability of the DRP models, we are exploring strategies that explicitly address the imbalance in the DRP datasets. We reframe the problem as a multi-objective optimization across multiple drugs to maximize deep learning model performance. We implement this approach by constructing Multi-Objective Optimization Regularized by Loss Entropy loss function and plugging it into a Deep Learning model. We demonstrate the utility of proposed drug discovery methods and make suggestions for further potential application of the work to achieve desirable outcomes in the healthcare field.

Biochemistry & Molecular Biology↗
DOE OSTI2025

Computing chemical potentials with machine-learning-accelerated simulations to accurately predict thermodynamic properties of molten salts

The successful design and deployment of next-generation nuclear technologies heavily rely on thermodynamic data for relevant molten salt systems. However, the lack of accurate force fields and efficient methods has limited the quality of thermodynamic predictions from atomistic simulations. Here we propose an efficient free energy framework for computing chemical potentials, which is the central free energy quantity behind many thermodynamic properties. We accelerate our simulations without sacrificing accuracy by using machine learning interatomic potentials trained on density functional theory (DFT) data. Using lithium chloride as our model system, we compute chemical potentials with DFT-accuracy for solid and liquid phases by transmuting ions into noninteracting particles. Notably, in the liquid phase, we demonstrate consistency whether we transmute one ion pair or the entire system into ideal gas particles. By locating the temperature where the chemical potential of solid and liquid phases cross, we predict a melting point of 880 ± 18 K for lithium chloride, which is remarkably close to the experimental value of 883 K. With this successful demonstration, we lay the foundation for high-throughput thermodynamic predictions of many properties that can be derived from the chemical potentials of the minority and majority components in molten salts.

Gibson, Luke D. [Oak Ridge National Laboratory (OR↗
DOE OSTI2024

Portable, heterogeneous ensemble workflows at scale using libEnsemble

libEnsemble is a Python-based toolkit for running dynamic ensembles, developed as part of the DOE Exascale Computing Project. The toolkit utilizes a unique generator–simulator–allocator paradigm, where generators produce input for simulators, simulators evaluate those inputs, and allocators decide whether and when a simulator or generator should be called. The generator steers the ensemble based on simulation results. Generators may, for example, apply methods for numerical optimization, machine learning, or statistical calibration. libEnsemble communicates between a manager and workers. Flexibility is provided through multiple manager–worker communication substrates each of which has different benefits. These include Python’s multiprocessing, mpi4py, and TCP. Multisite ensembles are supported using Balsam or Globus Compute. We overview the unique characteristics of libEnsemble as well as current and potential interoperability with other packages in the workflow ecosystem. We highlight libEnsemble’s dynamic resource features: libEnsemble can detect system resources, such as available nodes, cores, and GPUs, and assign these in a portable way. These features allow users to specify the number of processors and GPUs required for each simulation; and resources will be automatically assigned on a wide range of systems, including Frontier, Aurora, and Perlmutter. Such ensembles can include multiple simulation types, some using GPUs and others using only CPUs, sharing nodes for maximum efficiency. We also describe the benefits of libEnsemble’s generator–simulator coupling, which easily exposes to the user the ability to cancel, and portably kill, running simulations based on models that are updated with intermediate simulation output. We demonstrate libEnsemble’s capabilities, scalability, and scientific impact via a Gaussian process surrogate training problem for the longitudinal density profile at the exit of a plasma accelerator stage. In conclusion, the study uses gpCAM for the surrogate model and employs either Wake-T or WarpX simulations, highlighting efficient use of resources that can easily extend to exascale.

Dynamic ensembles↗
DOE OSTI2025

Machine Learned Empirical Numerical Integrator from Simulated Data

Recently, a number of state-of-the-art surrogate machine learning (ML) models have been designed for global weather and climate prediction, which have been trained using reanalysis data products. Reanalysis data products are constructed using numerical model simulations that combine numerical integration of partial differential equations and parameterization schemes. These products are typically only archived and made available using coarsened spatial and temporal resolutions. This study explores the impact of the numerical generation methods used to produce the training datasets and the temporal resolution of those datasets on machine learning surrogate models. Using the nonlinear vector autoregression (NVAR) machine as an explainable ML technique, simple dynamical systems are emulated with ML models trained on data produced by three classical numerical integration schemes. NVAR is validated as a skillful ML method, capable of producing accurate predictions and, more importantly, reconstructing both the underlying dynamics and the numerical integration scheme used to generate the training data. However, the machine fails to generalize predictions on unseen test data generated by different numerical integration schemes, despite the underlying dynamical system being the same. This result provides a word of caution for the growing field of machine learning emulation of weather and climate dynamics. Furthermore, we illustrate using NVAR that training on temporally coarsened data may increase the required complexity of ML models and potentially introduce new numerical challenges. Finally, we discover that empirical integration schemes with arbitrary time-stepping sizes can be constructed directly from the data, which implies a potential for the development of empirical numerical integration schemes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Deep Generative Models in Energy System Applications: Review, Challenges, and Future Directions

In recent years, with the advent of mature machine learning products like ChatGPT, Stable Diffusion, and Sora, the world has witnessed tremendous changes driven by the rapid development of generative artificial intelligence (GAI). Beyond applications in text, speech, image, and video creation, deep generative models (DGMs) underpinning these cutting-edge technologies have also been employed by domain researchers to address scientific and engineering challenges. This paper aims to fill a gap in the research community by providing a systematic review of how DGMs have been utilized in energy system applications. After introducing four most popular DGMs, we review and categorize 196 research articles into five focus areas: data generation, forecasting, situational awareness, modeling, and optimal decision-making. Through this classification, we uncover trends in how DGMs are employed for each type of problem, highlighting GAI techniques that contribute to breakthroughs over traditional methods. We discuss limitations in existing literature, engineering challenges, and propose future directions, all tailored to the unique nature of problems in energy system engineering. Our goal is to offer insights for energy system domain researchers, providing a comprehensive view of existing studies and potential future opportunities.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

CaloTrilogy: Toward a Breakthrough in One-Step, End-to-End, Physics-Guided Shower Generation for Modern Calorimeters

High-precision calorimeter simulation at current and future colliders imposes rapidly growing computational demands, motivating the development of machine-learning surrogates for traditional Monte Carlo tools such as Geant4. Flow matching and diffusion-based generative models have become leading approaches for high-dimensional fast simulation because of their sample quality, but typically require ${\cal O}(100)$ function evaluations at inference and often rely on auxiliary networks to constrain global observables, compromising streamlined end-to-end generation. We introduce a unified framework that improves the balance between speed, shower quality, and physics fidelity. The method combines: (i) an average velocity field integrator that enables sampling in one or a few evaluations; (ii) a learned generative prior in shower space, constructed from data rather than random noise; and (iii) physics-guided loss terms that impose inductive biases on key observables during training. These elements are training time regularizers, preserving end-to-end inference with no additional cost. With only one or a few evaluation steps, the model achieves shower quality competitive with state-of-the-art flow and diffusion approaches, tested on several public high granularity calorimeter datasets. The results demonstrate inter-layer shower structure consistent with the underlying physics, providing a strong candidate for future fast simulation workflows.

Jiang, Cheng [Edinburgh U.]↗
DOE OSTI2026

Generative models on phase space

Deep generative models such as diffusion and flow matching are powerful machine learning tools capable of learning and sampling from high-dimensional distributions. They are particularly useful when the training data appears to be concentrated on a submanifold of the data embedding space. For high-energy physics data, consisting of collections of relativistic energy-momentum 4-vectors, this submanifold can enforce extremely strong physically-motivated priors, such as energy and momentum conservation. If these constraints are learned only approximately, rather than exactly, this can inhibit the interpretability and reliability of such generative models. To remedy this deficiency, we introduce generative models which are, by construction, confined at every step of their sampling trajectory to the manifold of massless N-particle Lorentz-invariant phase space in the center-of-momentum frame. In the case of diffusion models, the "pure noise" forward process endpoint corresponds to the uniform distribution on phase space, which provides a clear starting point from which to identify how correlations among the particles emerge during the reverse (de-noising) process. We demonstrate that our models are able to learn both few-particle and many-particle distributions with various singularity structures, paving the way for future interpretability studies using generative models trained on simulated jet data.

Bogorad, Zachary [Fermilab]↗