Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “multi-fidelity modelling”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Scientific machine learning for closure models in multiscale problems: A review

Here, closure problems are omnipresent when simulating multiscale systems, where some quantities and processes cannot be fully prescribed despite their effects on the simulation's accuracy. Recently, scientific machine learning approaches have been proposed as a way to tackle the closure problem, combining traditional (physics-based) modeling with data-driven (machine-learned) techniques, typically through enriching differential equations with neural networks. This paper reviews the different reduced model forms, distinguished by the degree to which they include known physics, and the different objectives of a priori and a posteriori learning. The importance of adhering to physical laws (such as symmetries and conservation laws) in choosing the reduced model form and choosing the learning method is discussed. The effect of spatial and temporal discretization and recent trends toward discretization-invariant models are reviewed. In addition, we make the connections between closure problems and several other research disciplines: inverse problems, Mori-Zwanzig theory, and multi-fidelity methods. In conclusion, much progress has been made with scientific machine learning approaches for solving closure problems, but many challenges remain. In particular, the generalizability and interpretability of learned models is a major issue that needs to be addressed further.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Multi-Fidelity Bayesian Optimization with Gaussian Processes for Double Shell Inertial Confinement Fusion Target Design

Reliable, secure access to energy is a major focus for national security efforts. One potential route to such energy is through fusion reactions in inertial confinement fusion (ICF) experiments. Such experiments are carried out at facilities such as the National Ignition Facility (NIF) in Livermore, California, where high powered lasers are used to compress a DT fuel-containing target to the necessary high temperature, high pressure conditions. These experiments are limited in number, which creates a heavy dependence on high fidelity predictive physics simulations and analysis performed “pre shot,” or before the experiment occurs. Many of these simulations in higher dimensions (2D and 3D) are computationally expensive, so finding optimal simulation-based designs presents its own challenges. In this work, we present our multi-fidelity Bayesian optimization with Gaussian processes (GPs) for ICF double shell targets, where a 1D surrogate model is used to help find a 2D surrogate model, enabling us to find optimal targets in the higher fidelity (2D), while saving computational cost.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

A critical review on additive manufacturing of refractory alloys from a data analytics perspective- beyond nickel-based superalloys

Refractory alloys (RAs) are promising materials due to their exceptional physicochemical properties, but most research remains at the laboratory scale. For broader adoption, advancements in manufacturing are essential. Because their high stability makes conventional methods like machining and casting difficult, additive manufacturing (AM) is emerging as an effective approach for fabricating refractory alloy components. However, AM's repeated non-equilibrium thermal cycles introduce undesired features (e.g. defects, anisotropic microstructures, and residual stresses), which are magnified due to RAs’ unique properties. This paper comprehensively reviews the state-of-the-art methods of AM for refractory alloys. It explores data analytics techniques to establish design rules based on multi-fidelity experimental and computational methods. Furthermore, it investigates integrated, collaborative efforts to harmonise standalone databases, information, knowledge, and predictive models at multi-physics, multi-stage, and multi-scale. Unlike the existing literature that focuses primarily on material systems or process fundamentals, this work provides an integrated perspective on AM of refractory alloys from a data analytics standpoint, highlighting the roles of integrated computational materials engineering (ICME), verification, validation, and uncertainty quantification (VV&UQ), and digital twin-driven qualification in overcoming data scarcity and accelerating rapid qualification.

Additive manufacturing↗
DOE OSTI2023

Evaluation of Multi-Fidelity Soil Moisture Products Across the Continental United States

We have aggregated the most recent soil moisture datasets from a diverse range of sources, encompassing the Continental United States (CONUS). These sources encompass gridded data from remote sensing products, reanalysis products, machine learning-based projects, and land surface modeling products. Additionally, we have obtained and processed in-situ soil moisture observations from the International Soil Moisture Network. The collected datasets exhibit variations in both temporal and spatial resolutions. Among the 20 datasets, six are available at a spatial resolution of 0.25 degrees, while three are at a coarser spatial resolution of 25 km. To minimize spatial interpolation, we conducted data uncertainty evaluations at the 0.25-degree spatial resolution. For our data evaluations, we maintained a monthly temporal resolution, which effectively captures soil moisture seasonality and interannual variability. Our data processing strategy preserves the raw data and interpolated data at their original temporal resolutions. Datasets with higher temporal resolutions, including daily, three-hourly, and hourly datasets, are set aside for subsequent analyses. These analyses will delve into topics such as soil moisture changes and recovery during extreme weather events. Furthermore, we have processed auxiliary data to enhance our evaluation, leveraging tools such as Google Earth Engine. This includes incorporating topography data, land use land cover data, Köppen-Geiger climate classification, and more to provide a comprehensive assessment from multiple sources.

Li, Lingcheng↗
DOE OSTI2025

An evaluation of multi-fidelity methods for quantifying uncertainty in projections of ice-sheet mass change

Abstract. This study investigated the computational benefits of using multi-fidelity statistical estimation (MFSE) algorithms to quantify uncertainty in the mass change of Humboldt Glacier, Greenland, between 2007 and 2100 using a single climate change scenario. The goal of this study was to determine whether MFSE can use multiple models of varying cost and accuracy to reduce the computational cost of estimating the mean and variance of the projected mass change of a glacier. The problem size and complexity were chosen to reflect the challenges posed by future continental-scale studies while still facilitating a computationally feasible investigation of MFSE methods. When quantifying uncertainty introduced by a high-dimensional parameterization of the basal friction field, MFSE was able to reduce the mean-squared error in the estimates of the statistics by well over an order of magnitude when compared to a single-fidelity approach that only used the highest-fidelity model. This significant reduction in computational cost was achieved despite the low-fidelity models used being incapable of capturing the local features of the ice-flow fields predicted by the high-fidelity model. The MFSE algorithms were able to effectively leverage the high correlation between each model's predictions of mass change, which all responded similarly to perturbations in the model inputs. Consequently, our results suggest that MFSE could be highly useful for reducing the cost of computing continental-scale probabilistic projections of sea-level rise due to ice-sheet mass change.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Self-supervised and multi-fidelity learning for extended predictive soil spectroscopy

Infrared spectroscopy is a cost-effective, non-destructive, and environmentally benign technology that is increasingly recognized as an important solution for meeting the global demand for soil data. While both near-infrared (NIR) and mid-infrared (MIR) diffuse reflectance spectroscopy enable rapid estimation of soil properties, they present a significant trade-off: NIR offers superior scalability and lower operational costs, whereas MIR provides higher analytical fidelity by capturing fundamental molecular vibrations. In this study, we propose a self-supervised, multi-fidelity learning framework designed to bridge this gap. Our approach leverages large-scale MIR spectral libraries to learn a compact, transferable latent representation, into which NIR spectra are subsequently aligned for downstream prediction. The workflow consists of pretraining a latent model on a large MIR library, adapting the representation using a smaller paired NIR–MIR dataset, and evaluating generalization on an independent external test set. Across a range of chemical and physical soil properties, we found that MIR-derived embeddings improved prediction accuracy relative to baseline models that used raw MIR inputs. Predictions derived from the spectrum conversion (NIR to MIR) task did not match the performance of the original MIR spectra but were similar or superior to predictive performance of NIR-only models, suggesting the unified spectral latent space can effectively leverage the larger and more diverse MIR dataset for prediction of soil properties not well represented in current NIR libraries.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

FENIX: Towards a Fully Integrated Multiphysics Framework for Plasma Facing Component Modeling

Computational tools have a crucial role to play in accelerating the deployment of fusion as a clean, reliable, abundant, and sustainable energy source. Multiphysics, high-fidelity simulation capabilities can help model, study, and predict intricate interactions between materials performance, plasma exposure, neutron irradiation, and engineering processes. As such, they can assist in the resolution of scientific and engineering challenges underpinning design, construction, and commission of fusion power plants. To address these needs, ongoing efforts are leveraging the Multiphysics Object-Oriented Simulation Environment (MOOSE) framework and delivering new computational tools for the fusion community. These tools inherit crucial attributes from MOOSE. They are open-source, modular, integrated with nuclear industry-standard software quality assurance processes, and enable multiphysics, multi-fidelity, fully integrated, zero- to three-dimensional, and massively parallel simulations. After a short overview of these capabilities, we will present the development of Fusion ENergy Integrated multiphys-X (FENIX), a MOOSE-based application designed to enable plasma facing component design and performance evaluation. Throughout their lifetime, plasma facing components are exposed to extreme thermal loads, repeated thermal shocks, and irradiation by plasma ions, neutral particles, and high-energy neutrons. Consequently, designing a plasma facing component with acceptable lifetime degradation is extremely challenging. FENIX aims to model the multiphysics environment in which plasma facing components evolve to accelerate their design studies. To that end, FENIX couples existing MOOSE capabilities such as heat transfer, thermomechanics, and thermal hydraulics, with tritium transport via the MOOSE-based Tritium Migration Analysis Program, Version 8 (TMAP8), with neutronics via the MOOSE-based high-fidelity neutron-photon transport and fluid dynamics code Cardinal, and finally with Particle-in-Cell plasma simulation capabilities being developed in this project. In this study, we present the current FENIX capabilities and preliminary results of its application to model the Tritium Plasma Experiment set up at Idaho National Laboratory.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Multifidelity uncertainty quantification with models based on dissimilar parameters

Multifidelity uncertainty quantification (MF UQ) sampling approaches have been shown to significantly reduce the variance of statistical estimators while preserving the bias of the highest-fidelity model, provided that the low-fidelity models are well correlated. However, maintaining a high level of correlation can be challenging, especially when models depend on different input uncertain parameters, which drastically reduces the correlation. Existing MF UQ approaches do not adequately address this issue. In this work, we propose a new sampling strategy that exploits a shared space to improve the correlation among models with dissimilar parameterization. We achieve this by transforming the original coordinates onto an auxiliary manifold using the adaptive basis (AB) method (Tipireddy and Ghanem, 2014). The AB method has two main benefits: (1) it provides an effective tool to identify the low-dimensional manifold on which each model can be represented, and (2) it enables easy transformation of polynomial chaos representations from high- to low-dimensional spaces. This latter feature is used to identify a shared manifold among models without requiring additional evaluations. Here we present two algorithmic flavors of the new estimator to cover different analysis scenarios, including those with legacy and non-legacy high-fidelity (HF) data. We provide numerical results for analytical examples, a direct field acoustic test, and a finite element model of a nuclear fuel assembly. For all examples, we compare the proposed strategy against both single-fidelity and MF estimators based on the original model parameterization.

42 ENGINEERING↗
DOE OSTI2023

A PIC Bootstrapping Strategy for Exascale CFD-DEM Simulation

The exascale computing era comes with the release of MFIX-Exa, a new code for studying gas-particle fluidization using CFD-DEM and PIC models on massively parallel, heterogeneous high-performance computing architectures. However, as compute resources continue to grow in scale and complexity, so too does the impetus to use them efficiently. Here, we propose a novel bootstrapping method to minimize neglected simulation time and cast the approach more broadly among a growing field of multi-fidelity methods.

Porcu, Roberto↗
DOE OSTI2024

HydraGNN_Predictive_GFM_2024 - Ensemble of predictive graph foundation models for ground state atomistic materials modeling

We provide the ensemble of fifteen pre-trained graph foundation models (GFMs) for atomistic materials modeling applications. Each one of the fifteen GFMs has been trained on five open-source datasets that (once aggregated) amount to over 154 million atomistic structures, which cover over two-thirds of the natural elements of the periodic table and that comprises a broad set of organic and inorganic compounds. This vast set of atomistic structures comprises ground state configurations that are dynamically stable (i.e., equilibrated structures with atomic forces approximately close to zero values) as well as dynamically unstable structures (i.e., non-equilibrium structures with non-negligible non-zero values of atomic forces). The ensemble of datasets aggregated does NOT include excited states. The datasets have been curated to remove atomistic structures with spectral norm of the force tensor above 100 eV/angstrom. Moreover, a linear term of the energy was computed for each dataset using a linear regression model that uses the chemical concentration of each natural element as regressor. The linear term predicted by the linear regression model has been subtracted from each original energy value to perform a re-alignment of the energy values across different electronic structures approximation theories performed to generate the diverse multi-source, multi-fidelity datasets. The folder "ADIOS_files" 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 "ADIOS_files" directory contains 6 sub-directories named as follows: - ANI1x-v3.bp - MPTrj-v3.bp - OC2020-20M-v3.bp - OC2020-v3.bp - OC2022-v3.bp - qm7x-v3.bp Each sub-directory contains the pre-processed datasets converted in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used to the development, training, and performance testing of the ensemble go predictive graph foundation models. Each GFM was developed using HydraGNN (https://github.com/ORNL/HydraGNN) as underlying graph neural network (GNN) architecture. The multi-task learning (MTL) capability of HydraGNN was used to simultaneously train the GFMs on labeled values for direct predictions of energy (a total system property of an atomistic structure that measures the chemical stability) and atomic forces (an atomic level property of an atomistic structure that measures the dynamical stability). The hyper parameters of the GFM have been tuned using scalable hyperparameter optimization (HPO) algorithms implemented in the software DeepHyper (https://github.com/deephyper/deephyper). The pre-training of each HPO trial was performed using distributed data parallelism (DDP) to scale the training across 128 compute nodes of the exascale OLCF supercomputer Frontier. Each HPO trial was trained only for 10 epochs and an early stopping was performed to avoid wasting significant computational resources on GNN architectures that were clearly underperforming. For each HPO trial, the 'omnistat' tool developed by (AMD Research - Advanced Micro Device) was used to measure the total energy consumption in kWh. The ensemble of GFMs was obtained by selecting the fifteen best performing HPO trials. Four models have been selected for their clear advantage in accuracy, and these are the GFMs with IDs 229, 156, 147, 260. Additional eleven models have been selected based on judicious balance between accuracy and energy consumption needed for training, and these are the GFMs with IDs 165, 78, 137, 1, 175, 171, 181, 67, 179, 167, 351. Each selected GFM of the ensemble was continued to cumulate a total of at most 30 epochs. In some cases, the total number of epochs actually performed was les than 30 due to two combined factors: (1) the size of the GFM (i.e., the number of model parameters to train) and (2) the total wall-clock time for which the computational resources could be allocated on OLCF-Frontier. The "Ensemble_of_models" directory contains 15 sub-directories named as follows: - gfm_0.229 - gfm_0.156 - gfm_0.147 - gfm_0.260 - gfm_0.165 - gfm_0.78 - gfm_0.137 - gfm_0.1 - gfm_0.175 - gfm_0.171 - gfm_0.181 - gfm_0.67 - gfm_0.179 - gfm_0.167 - gfm_0.351 Each one of these sub-directories refers to one of the fifteen HPO trials that have been selected to continue the pre-training with at most 30 epochs. With each sub-directory associated with a specific HPO trial, the following files can be found: - config.json: file for argument parsing to develop and train an HydraGNN architecture - gfm_0.ID_epoch_N.pk: file with model parameters for HPO ID trial after N epochs of training The ensemble of fifteen GFM architectures was used for (1) ensemble averaging to stabilize the predictions of energy and atomic forces after pre-training for post-processing analysis and (2) ensemble uncertainty quantification (UQ). The code used to develop, pre-train, and load the pre-trained models for post-processing analysis is available on the ORNL-GitHub at the following link: https://github.com/ORNL/HydraGNN/tree/Predictive_GFM_2024

36 MATERIALS SCIENCE↗
DOE OSTI2025

Simulations of edge and SOL turbulence in diverted negative and positive triangularity plasmas

Optimizing the performance of magnetic confinement fusion devices is critical to achieving an attractive fusion reactor design. Negative triangularity (NT) scenarios have been shown to achieve excellent levels of energy confinement, while avoiding edge localized modes. Modeling turbulent transport in the edge and SOL is key in understanding the impact of NT on turbulence and extrapolating the results to future devices and regimes. Previous gyrokinetic turbulence studies have reported beneficial effects of NT across a broad range of parameters. However, most simulations have focused on the inner plasma region, neglecting the impact of NT on the outermost edge. In this work, we investigate the effect of NT in edge and scrape-off layer simulations, including the magnetic X-point and separatrix. For the first time, we employ a multi-fidelity approach, combining global, non-linear gyrokinetic simulations with drift-reduced fluid simulations, to gain a deeper understanding of the underlying physics at play. First-principles simulations using the GENE-X code demonstrate that in comparable NT and PT geometries, similar profiles are achieved, while the turbulent heat flux is reduced by more than 50% in NT. Comparisons with results from the drift-reduced fluid turbulence code GRILLIX suggest that the turbulence is driven by trapped electron modes. The parallel heat flux width on the divertor targets is reduced in NT, primarily due to a lower spreading factor S.

GENE-X↗
DOE OSTI2024

Enabling Scale-Up Through Multi-Fidelity Adaptive Computing

We present ideas from our ongoing work in adaptive computing - an optimization framework that allows us to strategically deploy various fidelity level experiments and simulations to guide decision making. The framework aims to enable uncertainty quantified scale-up of simulations and experiments, which causes increased complexity. A key feature of the framework is the integration of user-specified local model trustworthiness estimates. Adaptive sampling strategies allow us to optimally exploit the multiple fidelity level information and trustworthiness measures to arrive at the best decisions within a highly limited budget of objective function evaluations.

adaptive sampling↗
DOE OSTI2025

DeepHyper: A Python Package for Massively Parallel Hyperparameter Optimization in Machine Learning

Machine learning models are increasingly applied across scientific disciplines, yet their effectiveness often hinges on heuristic decisions—such as data transformations, training strategies, and model architectures—that are not learned by the models themselves. Automating the selection of these heuristics and analyzing their sensitivity is crucial for building robust and efficient learning workflows. DeepHyper addresses this challenge by democratizing hyperparameter optimization, providing accessible tools to streamline and enhance machine learning workflows from a laptop to the largest supercomputer in the world. Building on top of hyperparameter optimization, it unlocks new capabilities around ensembles of models for improved accuracy and uncertainty quantification. All of these organized around efficient parallel computing.

ensemble↗
DOE OSTI2024

Multi-Fidelity Scheme for Accelerating 4D Finite Element Analysis for Mircoreactors

Next-generation microreactors are currently being designed to be operated terrestrial and extraterrestrial for remote surface power production. These systems will provide an alternative source of carbon-free energy that is versatile and can be utilized for various applications. These applications of microreactors have prompted the usage of high-fidelity unstructured finite element (FE) based approaches to provide time-dependent solutions for multiple physics fields. Computing these high-fidelity (4-D) solutions for multiple design iterations, physics fields, and transient events requires an immense computational cost. These high-fidelity solutions are computational expensive and the cost can be reduced through the implementation of less accurate low-fidelity solutions. Unlike the current fleet of commercial nuclear reactors, these next-gen systems present challenges due to the material and physical limitations required. Due to these constraints, the brute force technique of parameterizing important system characteristics determines whether a design meets the project objective. A system such as a nuclear reactor could have thousands of design parameters that affect system performance. In order to analyze the entire parameter space of such a complex system would require millions of CPU hours and countless design iterations. This costly approach is not practical due to regulatory and budget limitations. In this proposal, an approach that utilizes hybrid high-fidelity and low-fidelity FE models to reduce the computational cost of evaluating these applications in 4-D will be presented. The accuracy of the high-fidelity model and the computational efficiency of the low-fidelity model are taken advantage of to produce a solution that closely resembles the full order high-fidelity solution. By utilizing an unconverged coarse FE mesh, operation limits such as temperature, structural loading, etc., can be evaluated in an accelerated fashion and then can be spatially interpolated onto a finer mesh. The resulting error arising from the coarse mesh can be mitigated with a discrepancy function that actively quantifies and corrects the error in the coarse solution. This discrepancy function can be periodically updated with high-fidelity calculations across the temporal domain thus requiring less iterations on the fine mesh. As a result, the computational cost can be reduced for the evaluation and design iteration of microreactors. The proposal for this research contains three sections: Section 2 provides a literature review, Section 3 outlines the methodology for the proposed multi-fidelity scheme, and Section 4 displays preliminary results of the proposed research.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

Inverse design of photonic surfaces via multi fidelity ensemble framework and femtosecond laser processing

We demonstrate a multi-fidelity (MF) machine learning ensemble framework for the inverse design of photonic surfaces, trained on a dataset of 11,759 samples that we fabricate using high throughput femtosecond laser processing. The MF ensemble combines an initial low fidelity model for generating design solutions, with a high fidelity model that refines these solutions through local optimization. The combined MF ensemble can generate multiple disparate sets of laser-processing parameters that can each produce the same target input spectral emissivity with high accuracy (root mean squared errors < 2%). SHapley Additive exPlanations analysis shows transparent model interpretability of the complex relationship between laser parameters and spectral emissivity. Finally, the MF ensemble is experimentally validated by fabricating and evaluating photonic surface designs that it generates for improved efficiency energy harvesting devices. Our approach provides a powerful tool for advancing the inverse design of photonic surfaces in energy harvesting applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Multi-fidelity Uncertainty Quantification for Homogenization Problems in Structure-Property Relationships from Crystal Plasticity Finite Elements

Crystal plasticity finite element method (CPFEM) has been an integrated computational materials engineering (ICME) workhorse to study materials behaviors and structure-property relationships for the last few decades. These relations are mappings from the microstructure space to the materials properties space. Due to the stochastic and random nature of microstructures, there is always some uncertainty associated with materials properties, for example, in homogenized stress-strain curves. For critical applications with strong reliability needs, it is often desirable to quantify the microstructure-induced uncertainty in the context of structure-property relationships. However, this uncertainty quantification (UQ) problem often incurs a large computational cost because many statistically equivalent representative volume elements (SERVEs) are needed. In this article, we apply a multi-level Monte Carlo (MLMC) method to CPFEM to study the uncertainty in stress-strain curves, given an ensemble of SERVEs at multiple mesh resolutions. By using the information at coarse meshes, we show that it is possible to approximate the response at fine meshes with a much reduced computational cost. We focus on problems where the model output is multi-dimensional, which requires us to track multiple quantities of interest (QoIs) at the same time. In conclusion, our numerical results show that MLMC can accelerate UQ tasks around 2.23x, compared to the classical Monte Carlo (MC) method, which is widely known as ensemble average in the CPFEM literature.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Cross-functional transferability in foundation machine learning interatomic potentials

The rapid development of foundation potentials (FPs) in machine learning interatomic potentials demonstrates the possibility for generalizable learning of the universal potential energy surface. The accuracy of FPs can be further improved by bridging the model from lower-fidelity datasets to high-fidelity ones. In this work, we analyze the challenge of this transfer learning (TL) problem within the CHGNet framework. We show that significant energy scale shifts and poor correlations between GGA and r 2 SCAN hinder cross-functional transferability. By benchmarking different TL approaches on the MP-r 2 SCAN dataset, we demonstrate the importance of elemental energy referencing in the TL of FPs. By comparing the scaling law with and without the pre-training on a low-fidelity dataset, we show that significant data efficiency can still be achieved through TL, even with a target dataset of sub-million structures. We highlight the importance of proper TL and multi-fidelity learning in creating next-generation FPs on high-fidelity data.

Huang, Xu [University of California, Berkeley, CA ↗
DOE OSTI2025

HydraGNN v4.0

The new version of HydraGNN v4.0 provides additional core capabilities, such as: Inclusion of multi-body atomistic cluster expansion MACE, polarizable atom interaction neural network PAINN, and equivariant principal neighborhood aggregation (PNAEq) among the message passing layers supported -Inclusion of graph transformers to directly model long-range interactions between nodes that are distant in the graph topology Integration of graph transformers with message passing layers by combining the graph embedding generated by the two mechanisms, which allows for an improved expressivity of the HydraGNN architecture Improved re-implementation of multi-task learning (MTL) to allow its use for stabilized training across imbalanced, multi-source, multi-fidelity data Introduction of multi-task parallelism, a newly proposed type of model parallelism specifically for MTL architectures, which allows to dispatch different output decoding heads to different GPU devices Integration of multi-task parallelism with pre-existing distributed data parallelism to enable a 2D parallelization for distributed training Improved portability of the distributed training across Intel GPUs, which has been testes on ALCF exascale supercomputer Aurora Inclusion of 2-level fine-grained energy profilers portable across NVIDIA, AMD, and Intel GPUs to monitor the power and energy consumption associated with different functions executed by the HydraGNN code during data pre-load and training Restructuring of previous examples and inclusion of new sets of examples to illustrate the download, preprocess, and training of HydraGNN models on new large-scale open-source datasets for atomistic materials modeling (e.g., Alexandria, Transition1x, OMat24, OMol25)

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗