Search NASA⌕ Search

SEARCH · Search NASA

Results for “scientific training”

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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Scientific machine learning for modeling and simulating complex fluids

The formulation of rheological constitutive equations—models that relate internal stresses and deformations in complex fluids—is a critical step in the engineering of systems involving soft materials. While data-driven models provide accessible alternatives to expensive first-principles models and less accurate empirical models in many engineering disciplines, the development of similar models for complex fluids has lagged. The diversity of techniques for characterizing non-Newtonian fluid dynamics creates a challenge for classical machine learning approaches, which require uniformly structured training data. Consequently, early machine-learning based constitutive equations have not been portable between different deformation protocols or mechanical observables. Here, we present a data-driven framework that resolves such issues, allowing rheologists to construct learnable models that incorporate essential physical information, while remaining agnostic to details regarding particular experimental protocols or flow kinematics. These scientific machine learning models incorporate a universal approximator within a materially objective tensorial constitutive framework. By construction, these models respect physical constraints, such as frame-invariance and tensor symmetry, required by continuum mechanics. We demonstrate that this framework facilitates the rapid discovery of accurate constitutive equations from limited data and that the learned models may be used to describe more kinematically complex flows. This inherent flexibility admits the application of these “digital fluid twins” to a range of material systems and engineering problems. We illustrate this flexibility by deploying a trained model within a multidimensional computational fluid dynamics simulation—a task that is not achievable using any previously developed data-driven rheological equation of state.

Science & Technology - Other Topics↗
DOE OSTI2022

Reduced diffusion and enhanced retention of multiple radionuclides from pore structure characterization of barrier materials for enhanced repository performance

Fluid flow and chemical transport in porous media are the macroscopic consequences of pore structure, which integrates geometry (e.g., pore size and surface area, pore-size distribution) and topology (e.g., pore connectivity). Low-permeability geological media whose pores are poorly interconnected will exhibit the characteristics of anomalous diffusion and sample size-dependent effective porosity, which will strongly impact long-term net diffusion and retention of radionuclides in geological repository settings involving different host rocks and barrier materials. A suite of innovative and complementary experimental approaches is utilized to study the microscopic pore structure and macroscopic fluid flow & chemical transport for a range of host rocks and barrier materials, in addition to standard clay minerals and reference rocks. With a particular focus on quantifying the presence and magnitude of “isolated” pores for a reduced effective porosity in low-permeability geomedia, the integrated methodologies for basic properties and pore structure characterization of these geomedia include X-ray diffraction, thin section petrography, grain size distribution, water immersion porosimetry after vacuum-pulling for full saturation, mercury intrusion porosimetry, nitrogen physisorption, scanning electron microscopy, X-ray computed tomography, and (ultra-)small angle neutron (X-ray) scattering. In addition, custom-designed gas diffusion, tracer recipe involving a range of anionic and cationic chemicals with subsequent analyses by laser ablation and inductively coupled plasma-mass spectrometry, along with batch sorption, column transport, and imbibition tests were conducted for coupled effects of pore structure and chemical retention/transport. From the perspectives of pore structure in conjunction with multiple and complementary approaches to examining a range of sample sizes under different observational scales, we find that the poor pore connectivity is prevalent in low-permeability media (mudstone and crystalline rock) that is related to geological processes (e.g., compaction, diagenesis and thermal maturation). For example, the deep and organic matter-rich mudstones have a much smaller effective porosity than the total porosity (as a result of poor pore connectivity) and associated diffusion coefficient, and the effective porosity & diffusion coefficients are also dependent upon the sample sizes used in the measurement. Similarly, most of the pore space in the shallow mudstone is also controlled by pore-throat diameters in the 5-50 nm range of intergranular pore types from its fine-grained nature, but with an overall good pore connectivity. However, the nm-sized pore space (physically pore-network architecture) and strong sorption capacities (chemical retention from clay minerals) of both shallow and deep mudstones lead to the synergistic retention of cationic radionuclides and their utilities as effective host rocks and barrier materials. Our unique approaches of studying how the micro-scale pore structure affect macro-scale fluid flow, diffusion & retention, and chemical transport produce improved mechanistic understanding, and realistic quantification, of diffusion and retention of typical radionuclides in a range of generic host rocks and barrier materials (clay/shale, salt, crystalline rock, and tuff), with the overall results leading to scientifically-based understanding of enhanced isolation (from both diffusion and retention) of radionuclides and improved confidence on the long-term performance of geological repository to store high-level radioactive wastes. In addition to the training of 25 undergraduates, graduates, and postdocs of UTA, the scientists (organizations) involved in performing this work (e.g., discussion, sample sharing, and operation of SANS and SAXS instruments) include Ed Matteo, Yifeng Wang, and Kristopher Kuhlman (Sandia National Laboratories), Jens Birkholzer, Liange Zheng, Tim Kneafsey, and Sharon Borglin (Lawrence Berkeley National Laboratory), Mavrik Zavarin (Lawrence Livermore National Laboratory), Yukio Tachi and Yuta Fukatsu (Japan Atomic Energy Agency), Mieke de Craen (Euridice, Belgium), Markus Bleuel (NIST), Wei-Ren Chen, Gergely Nagy, Changwoo Do, William Heller, Larry Anovitz, and Kenneth Littrell (ORNL), as well as Jan Illvsky, Ivan Kuzmenko, Ju-Sang Park and Jon Almers (ANL). Key deliverables include a total of 13 peer-reviewed journal articles (nine published and three under review), 23 presentations at scientific conferences (AAPG, AAPG Southwest Section, AGU, Asian Clay Conference, GSA, GSA South-Central Section, IHLRWM, InterPore, International Conference on Chemistry and Migration Behavior of Actinides and Fission Products in the Geosphere, International Conference on Coupled Processes in Fractured Geological Media: Observation, Modeling and Application), and academic institutions (UTA, New Mexico State University; University of Poitiers, France; University of Helsinki, Finland; Uppsala University, Sweden; Istanbul Technical University, Turkey) and other organizations (Andra, France; Posiva Oy, Finland).

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2021

Emerging Frameworks for Advancing Scientific Workflows Research, Development, and Education

Lightning talks of the Workflows in Support of Large-Scale Science (WORKS) workshop are a venue where the workflow community (researchers, developers, and users) can discuss work in progress, emerging technologies and frameworks, and training and education materials. This paper summarizes the WORKS 2021 lightning talks, which cover four broad topics: (i) libEnsemble, a Python library to coordinate the concurrent evaluation of dynamic ensembles of calculations; (ii) Edu WRENCH, a set of online pedagogic modules that provides simulation-driven hands-on activity in the browser; (iii) VisDict, an envisioned visual dictionary framework that will translate terms, jargon, and concepts between research domains and workflow providers; and (iv) Pegasus Kickstart, a lightweight tool for capturing workflow tasks' performance, including performance metrics from Nvidia GPUs.

Casanova, Henri↗
DOE OSTI2026

HydraGNN_Predictive_GFM_2026 - Ensemble of predictive graph foundation models for atomistic materials modeling

This release contains data and parameters of HydraGNN-based graph foundation models trained as a result of the work published in the pre-print "Exascale Multi-Task Graph Foundation Models for Imbalanced, Multi-Fidelity Atomistic Data" by M. Lupo Pasini et al. (https://arxiv.org/abs/2604.15380). We jointly train on 16 open first-principles datasets (544+ million structures covering 85+ elements) using a multi-task architecture with per-dataset heads and a scalable ADIOS2/DDStore data pipeline. On Frontier, we execute six large-scale DeepHyper hyperparameter optimization campaigns in FP64 and promote the top-performing message-passing models to sustained 2,048-node training, yielding a PaiNN-based lead model. The version of HydraGNN used to generate the outputs provided in this release is HydraGNN v5.0 (https://github.com/ORNL/HydraGNN/releases/tag/v5.0) The list of datasets used for the training of the graph foundation model is the following: 1) Alexandria [1] 2) ANI1x [2] 3) MPTrj [3] 4) Open Catalyst 2020 (OC20) [4] 5) Open Catalyst 2022 (OC22) [5] 6) Open Catalyst 2025 (OC25) [6] 7) Open Direct ir Capture 2023 (ODAC23) [7] 8) Open Materials 2024 (OMat24) [8] 9) Open Molecules 2025 (OMol25) [9] 10) OMol25-neutral (subset of OMol25 that contains only molecules with zero total charge) 11) OMol25-non-neutral (subset of OMol25 that contains only molecules with non-zero total charge) 12) Open Polymers 2026 (OPoly2026) [10] 13) Nabla2DFT [11] 14) QCML [12] 15) QM7X [reference 13] 16) transition1x [14] Dataset references: [1] J. Schmidt et al., “A dataset of 175k stable and metastable materials calculated with the PBEsol and SCAN functionals,” Scientific Data, vol. 9, p. 64, 2022. [2] J. S. Smith et al., “The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules,” Scientific Data, vol. 7, p. 134, 2020. [Online]. Available: https: //www.nature.com/articles/s41597-020-0473-z [3] A. Jain et al., “Commentary: The Materials Project: A materials genome approach to accelerating materials innovation,” APL Materials, vol. 1, no. 1, p. 011002, 07 2013. [Online]. Available: https://doi.org/10.1063/1.4812323 [4] L. Chanussot et al., “Open catalyst 2020 (oc20) dataset and community challenges,” ACS Catalysis, vol. 11, no. 10, pp. 6059–6072, 2021. [Online]. Available: https://doi.org/10.1021/acscatal.0c04525 [5] K. Tran et al., “Open catalyst 2022 (oc22) dataset and challenges for oxidation electrocatalysts,” ACS Catalysis, vol. 13, no. 5, pp. 3066–3084, 2023. [Online]. Available: https://doi.org/10.1021/acscatal.2c05426 [6] S. J. Sahoo et al., “The open catalyst 2025 (oc25) dataset and models for solid-liquid interfaces,” arXiv preprint arXiv:2509.17862, 2025. [Online]. Available: https://arxiv.org/abs/2509.17862 [7] A. Sriram et al., “The open DAC 2023 dataset and challenges for sorbent discovery in direct air capture,” ACS Central Science, vol. 10, no. 5, pp. 923–941, 2024. [8] L. Barroso-Luque et al., “Open materials 2024 (omat24) inorganic materials dataset and models,” 2024. [Online]. Available: https://arxiv.org/abs/2410.12771 [9] D. S. Levine et al., “The open molecules 2025 (OMol25) dataset, evaluations, and models,” 2025. [Online]. Available: https://arxiv.org/abs/2505.08762 [10] D. S. Levine et al., The open polymers 2026 (OPoly26) dataset and evaluations,” arXiv preprint arXiv:2512.23117, 2025. [Online]. Available: https://arxiv.org/abs/2512.23117 [11] K. Khrabrov et al., “Nabla2dft: A universal quantum chemistry dataset of drug-like molecules and a benchmark for neural network potentials,” in NeurIPS 2024 Datasets and Benchmarks Track, 2024. [Online]. Available: https://openreview.net/forum?id=ElUrNM9U8c [12] S. Ganscha et al., “The QCML dataset, quantum chemistry reference data from 33.5M DFT and 14.7B semi-empirical calculations,” Scientific Data, vol. 12, p. 406, 2025. [13] J. Hoja et al., “QM7-X, a comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules,” Scientific Data, vol. 8, p. 43, 2021. [Online]. Available: https://www.nature.com/articles/s41597-021-00812-2 [14] M. Schreiner et al., “Transition1x - a dataset for building generalizable reactive machine learning potentials,” Scientific Data, vol. 9, p. 779, 2022. The folder "datasets_ADIOS2_format" contains the set of pre-processed datasets in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used for the development and training of GFMs in this work. The "datasets_ADIOS2_format" directory contains 2 sub-directories, one for the version "v1" of the datasets and one for the version "v2" of the datasets. The version "v1" of the datasets provides values of the total energy as they are extracted from the original data as it was released by the respective institutions. The version "v2" of the datasets provides values of the energy that have been realigned. The realignment was performed by training a linear regression model that predicts the total energy as a function of the chemical composition of the atomistic structure, and then subtract such prediction from the original value of the total energy. Both folders "v1" and "v2" contain 16 sub-directories, each corresponding to an ADIOS2-formatted dataset The folder "DeepHyper-results" contains the configurational files and model's parameters for all the 186 HPO trials that were successfully completed by the scalable hyperparameter optimization (HPO) runs on Frontier. The content of the folder "DeepHyper-results" I structured as follows: 1) task-list.txt: list of mpnn name, jobid, and deephyper task id 2) gfm_${MPNN}_${JOBID}_0.${TASKID}: run directory with checkpoint files 3) gfm_${MPNN}: deephyper summary directory (*.csv) for each specific MPNN type 4) deephyper-experiment-${JOBID}: output and error logs for each job The file "deephyper-sorted.csv" contains the details of each HydraGNN model built and tested by HPO, obtained by merging the (*.csv) filed from each HPO run executed. Out of all the HPO trials, we selected 10 to continue the training of the respective HydraGNN models. Due to limited computational budget available in the LRN070 allocation we could not complete the training till convergence for all these 10 selected models. The folder "models" contains multiple sub-folders, one per each HydraGNN model trained. Each model sub-folder contains the parameters of each HydraGNN model, with multiple checkpoint-restarts. The list of sub-folders are as follows: 1) multidataset_hpo-BEST1-fp64 2) multidataset_hpo-BEST2-fp64 3) multidataset_hpo-BEST3-fp64 4) multidataset_hpo-BEST4-fp64 5) multidataset_hpo-BEST5-fp64 6) multidataset_hpo-BEST6-fp64 7) multidataset_hpo-BEST7-fp64 8) multidataset_hpo-BEST8-fp64 9) multidataset_hpo-BEST9-fp64 10) multidataset_hpo-BEST10-fp64 Within each one of these folders, additional auxiliary log files are provided with descriptions about how the training proceeded. The lead PaiNN-model is contained inside "multidataset_hpo-BEST6-fp64". The file "mlp_branch_weights" contains the parameters of the multi-layer perceptron (MLP) used to reconcile the predictions of the 16 output decoding heads of the HydragNN architectures. The MLP takes in input the chemical composition of the atomistic structure and predicts averaging weights to linearly mix the predictions of each output decoding head toward consolidating them into a single one. The folder "1.1billion-structure-inference" contains 1.1 billion atomistic structures randomly generated. Each structures is associated with energy and forces predicted with the lead-PaiNN model combined with the MLP model for reconciliation of the multi-branch predictions generated by the 16 output decoding heads. The folder "1.1billion-structure-inference" contains 9,300 (*.tar.gz) subdirectories, one per Frontier compute node used to execute the inference at exascale. Once uncompressed, each (*.tar.gz) subdirectory contains an ADIOS2 (*.bp) file container, where each atomistic structure is stored as a PyTorch-Geometric Data object. The file "export_dataset_environment_variables.sh" contains the environment variables that need to be set before running the HydraGNN code to reproduce the results provided in this dataset release. The code that can be used to load the ADIOS2 files, load HydraGNN models, and run inference is available at: https://github.com/ORNL/HydraGNN/releases/tag/v5.0

36 MATERIALS SCIENCE↗
DOE OSTI2025

SPIKANs: separable physics-informed Kolmogorov–Arnold networks

Physics-Informed Neural Networks (PINNs) have emerged as a promising method for solving partial differential equations (PDEs) in scientific computing. While PINNs typically use multilayer perceptrons (MLPs) as their underlying architecture, recent advancements have explored alternative neural network structures. One such innovation is the Kolmogorov–Arnold Network (KAN), which has demonstrated benefits over traditional MLPs, including faster neural scaling and better interpretability. The application of KANs to physics-informed learning has led to the development of Physics-Informed KANs (PIKANs), enabling the use of KANs to solve PDEs. However, despite their advantages, KANs often suffer from slower training speeds, particularly in higher-dimensional problems where the number of collocation points grows exponentially with the dimensionality of the system. To address this challenge, we introduce Separable Physics-Informed Kolmogorov–Arnold Networks (SPIKANs). This novel architecture applies the principle of separation of variables to PIKANs, decomposing the problem such that each dimension is handled by an individual KAN. This approach drastically reduces the computational complexity of training without sacrificing accuracy, facilitating their application to higher-dimensional PDEs. Through a series of benchmark problems, we demonstrate the effectiveness of SPIKANs, showcasing their superior scalability and performance compared to PIKANs and highlighting their potential for solving complex, high-dimensional PDEs in scientific computing.

Kolmogorov-Arnold networks↗
DOE OSTI2024

Exact enforcement of temporal continuity in sequential physics-informed neural networks

The use of deep learning methods in scientific computing represents a potential paradigm shift in engineering problem solving. One of the most prominent developments is Physics-Informed Neural Networks (PINNs), in which neural networks are trained to satisfy partial differential equations (PDEs). While this method shows promise, the standard version has been shown to struggle in accurately predicting the dynamic behavior of time-dependent problems. To address this challenge, methods have been proposed that decompose the time domain into multiple segments, employing a distinct neural network in each segment and directly incorporating continuity between them in the loss function of the minimization problem. In this work we introduce a method to exactly enforce continuity between successive time segments via a solution ansatz. This hard constrained sequential PINN (HCS-PINN) method is simple to implement and eliminates the need for any loss terms associated with temporal continuity. The method is tested for a number of benchmark problems involving both linear and non-linear PDEs. Examples include various first order time dependent problems in which traditional PINNs struggle, namely advection, Allen–Cahn, and Korteweg–de Vries equations. Furthermore, second and third order time-dependent problems are demonstrated via wave and Jerky dynamics examples, respectively. Notably, the Jerky dynamics problem is chaotic, making the problem especially sensitive to temporal accuracy. Finally, the numerical experiments conducted with the proposed method demonstrated superior convergence and accuracy over both traditional PINNs and the soft-constrained counterparts.

42 ENGINEERING↗
DOE OSTI2023

Streaming Data from Experimental Facilities to Supercomputers for Real-Time Data Processing

In this paper we demonstrate direct data streaming from instruments and detectors at a large-scale experimental facility to a supercomputer for real-time data processing and feedback. Streaming data to supercomputers introduces the potential for novel scientific applications and workflow models, including the ability to provide real-time feedback from very large datasets during an experiment and the integration of real-time ML training and inference at scale. We discuss a successful demonstration for real-time processing of data from the Advanced Photon Source (APS) on the Polaris supercomputer using an EPICS-based streaming framework. We describe the capabilities of the streaming framework itself, and outline the architecture that allows us to process experimentally derived data on a supercomputer without file-based data transfers. We present throughput measurements that are indicative of system performance capable of sustaining the expected data production rates of the facility, as well as discuss some outstanding challenges and our future directions.

real-time processing↗
DOE OSTI2022

Corrosion of U233-Doped Uranium Oxide using Microfluidics Methods

The aim of work this year has been to investigate the role of alpha (α)-radiation and hydrogen (H 2 ) on the corrosion of uranium oxide (UO 2 ) using a microfluidic device. The microfluidic device, termed the Particle-Attached Microfluidic Electrochemical Cell, (PAMEC), enables monitoring of the UO 2 electrochemical corrosion potential (E corr ) under different environments, including de-aerated conditions and in the presence of dissolved H 2 . The Si 3 N 4 window allows us to study morphological and chemical changes under an electron microscope. We have previously demonstrated that the PAMEC matches the results from bulk electrochemical tests with UO 2 [1]. We used the high specific activity uranium (U) isotope, 233 U, (t 1/2 = 160,000 years) incorporated into UO 2 , to generate a localized a-field. The objective of the experiments were to mimic the radiation environment that would be experience at the surface of aged spent nuclear fuel (SNF) during long-term geologic disposal under anoxic conditions. Wittman et al. [2] predicted that in the presence of a pure a-radiation field and under H 2 conditions, the concentration of the radiolytic oxidant H 2 O 2 would be suppressed or even eliminated. In a UO 2 corrosion experiment this would be exhibited through a lowering of the measured UO 2 corrosion potential compared to identical conditions in the absence of dissolved H 2 . We found that the predictions of Wittman and co-workers were supported and that the corrosion potential of the 233 U-doped UO 2 in solution lowered with presence of H 2 gas and increased in the absence of H 2 , when under anoxic conditions. The PAMEC experiments indicate that H 2 O 2 has been eliminated in a solution sparged with H 2 while exposed to an a-radiation field. The corrosion potential of the a-doped 233 U(10%)- 238 UO 2 in a solution sparged with Ar/H 2 matched the corrosion potential of 238 UO 2 in solution sparged with air. This clearly demonstrated that the H 2 O 2 had been eliminated and that the only oxidant present was O 2 in this system in complete agreement with the modeling results of Wittman et al. These results point to the need to improve the Fuel Matrix Degradation (FMD) Process Model training data set that is being used in the FMD surrogate model that is being developed for the repository program. The incorporation of realistic radiation chemistry will improve the scientific basis for the FMD Model.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2023

Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces

Typical generative diffusion models rely on a Gaussian diffusion process for training the backward transformations, which can then be used to generate samples from Gaussian noise. However, real world data often takes place in discrete-state spaces, including many scientific applications. Here, we develop a theoretical formulation for arbitrary discrete-state Markov processes in the forward diffusion process using exact (as opposed to variational) analysis. We relate the theory to the existing continuous-state Gaussian diffusion as well as other approaches to discrete diffusion, and identify the corresponding reverse-time stochastic process and score function in the continuous-time setting, and the reverse-time mapping in the discrete-time setting. As an example of this framework, we introduce “Blackout Diffusion”, which learns to produce samples from an empty image instead of from noise. Numerical experiments on the CIFAR-10, Binarized MNIST, and CelebA datasets confirm the feasibility of our approach. Generalizing from specific (Gaussian) forward processes to discrete-state processes without a variational approximation sheds light on how to interpret diffusion models, which we discuss.

Santos, Javier E.↗
DOE OSTI2021

Interdisciplinary Research Collaborative Trains Students to See Through Turbulent Systems

Despite the availability of interdisciplinary academic training programs, the practice of environmental science is often hampered by a lack of convergence across diverse disciplines. This gap is particularly salient in settings characterized by complex environmental issues, such as multiple-use coastal ecosystems. In response, we developed and implemented a training, research, and communication framework to provide undergraduates with an authentic operative experience working at the interface of interdisciplinary science and public decision-making within a case study of marine renewable energy. In our program, students gained hands-on experience with the scientific process and learned how to make information relevant, useful, and accessible to diverse stakeholder groups. Application of this framework demonstrates that the process of integrating data from biological (visual and acoustic monitoring of fish and marine mammals), physical (hydrodynamics), and social (local ecological knowledge) sciences can provide a more complete understanding of complex and turbulent ecosystems for better informed decision-making. We offer several recommendations to facilitate the adaptation and implementation of our interdisciplinary framework to diverse research contexts, with a focus on interdisciplinary training for the next generation of marine scientists.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

NREL Stratus - Enabling Workflows to Fuse Data Streams, Modeling, Simulation, and Machine Learning

Integrating cloud services into advanced computing facilities provides significant new capabilities over focusing solely on traditional high performance computing (HPC) workloads. This brings complementary capabilities as well as enabling new focused roles for HPC. They are especially potent for workflows that fuse data streams, modeling and simulation ('modsim') and machine learning. A key challenge to adopting a hybrid edge-cloud-HPC model is to align optimal capability, data, and user intent on the right resources for each step in a workflow.?The NREL Stratus service provides a basis for this: Stratus layers capabilities needed to make?cloud services accessible to a lab-based scientific community on commercial offerings, and; currently supports upwards of 200 projects ranging from IOT integration to traditional modeling and simulation. This provides a real-world inventory of scientific workflow elements. A growing knowledge base enables placing these elements appropriately between the edge, cloud, and traditional HPC. This paper outlines a vision via reference architecture and the application of that architecture in a typical workflow highlighting multiple components: sensor data intake, cleaning and transforming (edge/cloud suitable); generation of synthetic data through modsim, computationally heavy ML training and hyperparameter optimization (HPC suitable), and; inference and deployment (cloud ideal). Every step in such a workflow involves a cost-benefit analysis regarding the data movement, computational efficiency, availability, latency, and resource capabilities. The reference architecture and examples outlined allow for understanding new opportunities in the context of emerging workflows that combine IOT, cloud, and HPC to bolster scientific productivity.

AI↗
DOE OSTI2022

28 NREL Stratus - Enabling Workflows to Fuse Data Streams, Modeling, Simulation, and Machine Learning: Preprint

Integrating cloud services into advanced computing facilities provides significant new capabilities over focusing solely on traditional high performance computing (HPC) workloads. This brings complementary capabilities as well as enabling new focused roles for HPC. They are especially potent for workflows that fuse data streams, modeling and simulation ('modsim') and machine learning. A key challenge to adopting a hybrid edge-cloud-HPC model is to align optimal capability, data, and user intent on the right resources for each step in a workflow.?The NREL Stratus service provides a basis for this: Stratus layers capabilities needed to make?cloud services accessible to a lab-based scientific community on commercial offerings, and; currently supports upwards of 200 projects ranging from IOT integration to traditional modeling and simulation. This provides a real-world inventory of scientific workflow elements. A growing knowledge base enables placing these elements appropriately between the edge, cloud, and traditional HPC. This paper outlines a vision via reference architecture and the application of that architecture in a typical workflow highlighting multiple components: sensor data intake, cleaning and transforming (edge/cloud suitable); generation of synthetic data through modsim, computationally heavy ML training and hyperparameter optimization (HPC suitable), and; inference and deployment (cloud ideal). Every step in such a workflow involves a cost-benefit analysis regarding the data movement, computational efficiency, availability, latency, and resource capabilities. The reference architecture and examples outlined allow for understanding new opportunities in the context of emerging workflows that combine IOT, cloud, and HPC to bolster scientific productivity.

AI↗
DOE OSTI2023

Code for the manuscript titled "Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces"

We would like to disclose two scripts, written in Jupyter notebook, in which we implement the "Blackout Diffusion Process" described in the manuscript "Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces" (LA-UR-23-20509), to be submitted to the International Conference of Machine Learning (ICML). The abstract of the manuscript is append below. == Typical generative diffusion models rely on a Gaussian diffusion process for training the backward transformations, which can then be used generate samples from Gaussian noise. However, real world data often takes place in discrete-state spaces, which occur in many scientific applications. Here, we develop a theoretical formulation for arbitrary discrete-state Markov processes in the forward diffusion process. We relate the theory to the existing continuous-state Gaussian diffusion and identify the corresponding reverse-time stochastic process and score function in the continuous-time setting, and the reverse-time mapping in the discrete-time setting. As an example of this framework, we introduce "Blackout Diffusion", which learns to produce samples from an empty image instead of from noise. Numerical experiments on the CIFAR-10 dataset confirm the feasibility of generative diffusion modeling in a discrete space. Generalizing from specific (Gaussian) forward processes to a more general framework also sheds light on how to interpret generative diffusion models and their mathematical structure, which we comment on.

Lin, Yen Ting↗
DOE OSTI2023

High-bandwidth image-based predictive laser stabilization via optimized Fourier filters

Controlling the delivery of kHz-class pulsed lasers is of interest in a variety of industrial and scientific applications, from next-generation laser-plasma acceleration to laser-based x-ray emission and high-precision manufacturing. The transverse position of the laser pulse train on the application target is often subject to fluctuations by external drivers (e.g., room cooling and heating systems, motorized optics stages and mounts, vacuum systems, chillers, and/or ground vibrations). For typical situations where the disturbance spectrum exhibits discrete peaks on top of a broad-bandwidth lower-frequency background, traditional PID (proportional-integral-derivative) controllers may struggle, since as a general rule PID controllers can be used to suppress vibrations up to only about 5%–10% of the sampling frequency. Here, a predictive feed-forward algorithm is presented that significantly enhances the stabilization bandwidth in such laser systems (up to the Nyquist limit at half the sampling frequency) by online identification and filtering of one or a few discrete frequencies using optimized Fourier filters. Furthermore, the system architecture demonstrated here uses off-the-shelf CMOS cameras and piezo-electric actuated mirrors connected to a standard PC to process the alignment images and implement the algorithm. To avoid high-end, high-cost components, a machine-learning-based model of the piezo mirror’s dynamics was integrated into the system, which enables high-precision positioning by compensating for hysteresis and other hardware-induced effects. A successful demonstration of the method was performed on a 1 kHz laser pulse train, where externally-induced vibrations of up to 400 Hz were attenuated by a factor of five, far exceeding what could be done with a standard PID scheme.

Natal, Joseph↗
DOE OSTI2021

HYPPO: A Surrogate-Based Multi-Level Parallelism Tool for Hyperparameter Optimization

We present a new software, HYPPO, that enables the automatic tuning of hyperparameters of various deep learning (DL) models. Unlike other hyperparameter optimization (HPO) methods, HYPPO uses adaptive surrogate models and directly accounts for uncertainty in model predictions to find accurate and reliable models that make robust predictions. Using asynchronous nested parallelism, we are able to significantly alleviate the computational burden of training complex architectures and quantifying the uncertainty. HYPPO is implemented in Python and can be used with both TensorFlow and PyTorch libraries. We demonstrate various software features on time-series prediction and image classification problems as well as a scientific application in computed tomography image reconstruction. Finally, we show that (1) we can reduce by an order of magnitude the number of evaluations necessary to find the most optimal region in the hyperparameter space and (2) we can reduce by two orders of magnitude the throughput for such HPO process to complete.

adaptation models↗
DOE OSTI2025

New Particle Formation and Growth in the Houston Atmosphere During TRACER (Final Report)

From 2020-2025, researchers from UC Irvine, UC Riverside, and Colorado State University collaborated on a Department of Energy-funded project to understand how airborne particles form and grow in urban atmospheres, conducting an intensive field campaign in Houston, Texas during summer 2022. Using advanced instruments to measure gas-phase chemicals, particle composition, and a specialized chamber to study particle growth, the team discovered that sulfur-containing compounds from industrial and power plant emissions are the dominant driver of new particle formation in Houston, with particles typically forming locally in the city and growing as air moves away in the urban plume. The research revealed an important methodological insight: measurements from fixed ground stations can be misleading when interpreting how particles actually evolve as air masses move, which has significant implications for how scientists worldwide interpret atmospheric observations. These findings improve understanding of urban air quality and help reduce uncertainties in climate models, since these particles play critical roles in cloud formation and Earth's radiation balance, while also providing detailed information about ultrafine particle composition relevant to public health. The project trained three doctoral students, developed enhanced computer models for urban particle formation, and made all data publicly available through the DOE Atmospheric Radiation Measurement data archive for use by the broader scientific community.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Radiation biology workforce in the United States

In recent decades, the principal goals of participants in the field of radiation biologists have included defining dose thresholds for cancer and non-cancer endpoints to be used by regulators, clinicians and industry, as well as informing on best practice radiation utilization and protection applications. Importantly, much of this work has required an intimate relationship between “bench” radiation biology scientists and their target audiences (such as physicists, medical practitioners and epidemiologists) in order to ensure that the requisite gaps in knowledge are adequately addressed. However, despite the growing risk for public exposure to higher-than-background levels of radiation, e.g. from long-distance travel, the increasing use of ionizing radiation during medical procedures, the threat from geopolitical instability, and so forth, there has been a dramatic decline in the number of qualified radiation biologists in the U.S. Contributing factors are thought to include the loss of applicable training programs, loss of jobs, and declining opportunities for advancement. This report was undertaken in order to begin addressing this situation since inaction may threaten the viability of radiation biology as a scientific discipline.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2022

MULTISTEP AND CONTINUOUS PHYSICS-INFORMED NEURAL NETWORK METHODS FOR LEARNING GOVERNING EQUATIONS AND CONSTITUTIVE RELATIONS

In this work, we investigate the applicability and relative merit of discrete and continuous versions of physics-informed neural network (PINN) methods for learning unknown governing equations or constitutive relations in a nonlinear dynamical system. In the case of unknown dynamics, entire right-hand-side (RHS) equations of the ordinary differential equations are unknown. In the case of unknown constitutive relations, however, the RHS equations are known up to the specification of constitutive relations (that may depend on the state of the system). We use a deep neural network to model unknown governing equations or constitutive relations. The discrete PINN approach combines classical multistep discretization methods for dynamical systems with neural-network-based machine learning methods. On the other hand, the continuous versions utilize deep neural networks to minimize the residual function for the continuous governing equations. We use the case of a fedbatch bioreactor system to study the effectiveness of these approaches and discuss conditions for their applicability. Our results indicate that the accuracy of the trained neural network models is much higher for the cases where we only have to learn a constitutive relation instead of all dynamics. This finding corroborates the well-known fact from scientific computing that building as much structural information as is available into an algorithm can enhance its efficiency and/or accuracy.

97 MATHEMATICS AND COMPUTING↗