Search NASA⌕ Search

SEARCH · Search NASA

Results for “Contrastive 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 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Aerosol Influences on Cloud Water: Insights From ARM EPCAPE Observations With Explainable Machine Learning

This study employs an explainable machine learning (ML) framework (XGBoost-SHapley Additive exPlanations analysis) to investigate controlling factors on cloud liquid water path (LWP) using EPCAPE observations near the California coast. Aerosols are found to be the dominant factor explaining LWP variability, surpassing meteorological factors (MFs). By isolating aerosol effects from meteorological influences, the ML reveals a negative linear relationship between LWP and cloud droplet number concentration (Nd) in log space, likely driven by entrainment drying via evaporation-entrainment feedback. This aligns with the negative regime of the inverted-V relationship reported in previous studies, while no positive LWP responses are found due to a limited number of precipitating cases in EPCAPE. Furthermore, the sensitivity of LWP to Nd shows a non-linear dependence on MFs like moisture contrast between surface and free troposphere and lower-tropospheric stability. This occurs due to the interplay between the MFs' direct effects on entrainment drying and indirect effects through LWP adjustments.

EPCAPE observations↗
DOE OSTI2021

Information-Theoretic Bounds on Quantum Advantage in Machine Learning

We study the performance of classical and quantum machine learning (ML) models in predicting outcomes of physical experiments. The experiments depend on an input parameter x and involve execution of a (possibly unknown) quantum process E. Our figure of merit is the number of runs of E required to achieve a desired prediction performance. We consider classical ML models that perform a measurement and record the classical outcome after each run of E, and quantum ML models that can access E coherently to acquire quantum data; the classical or quantum data are then used to predict the outcomes of future experiments. We prove that for any input distribution D(x), a classical ML model can provide accurate predictions on average by accessing E a number of times comparable to the optimal quantum ML model. In contrast, for achieving an accurate prediction on all inputs, we prove that the exponential quantum advantage is possible. For example, to predict the expectations of all Pauli observables in an n-qubit system ρ, classical ML models require 2 Ω(n) copies of ρ, but we present a quantum ML model using only O(n) copies. Our results clarify where the quantum advantage is possible and highlight the potential for classical ML models to address challenging quantum problems in physics and chemistry.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Time-resolved spray characterization via unified optical flow and binarization technique

This work leverages an unsupervised machine learning and advanced image processing techniques to characterize the breakup of fuel sprays in a small-scale combustor under reacting conditions, providing valuable insights into near-nozzle flow phenomenology. The proposed methodology integrates an improved optical flow model on a convolutional neural network to extract flow vectors with a binarization technique to assess droplets’ size and shape across the region of interest. The velocimetry approach demonstrates superior performance compared to a state-of-the-art optical flow model when applied to high-speed X-ray phase contrast spray images, achieving more accurate and reliable flow predictions. Moreover, breakup processes are quantified by breakup length and sphericity in accordance with velocity estimations, allowing a more complete characterization of the flow. This study establishes a robust methodology for analyzing spray morphology and primary breakup in compact combustors, contributing valuable means of understanding and optimizing fuel spray behavior in advanced combustion systems.

42 ENGINEERING↗
DOE OSTI2022

Dynamic calibration of differential equations using machine learning, with application to turbulence models

We present a methodology for calibration of parametric ordinary and partial differential equation models, using off-the-shelf software for back-propagation in Neural Networks (NN). As a prototypical example, we consider calibration of a Reynolds-averaged Navier-Stokes (RANS) turbulence closure model, against ground truth data from direct numerical simulations (DNS) of two different turbulent flows. Numerical time integration is represented as a custom NN, where only the RANS model parameters are trainable. A loss function is defined to quantify the mismatch between the NN prediction and the ground truth over a predefined, finite time integration window. This loss function is then minimized using a gradient descent method utilizing the back-propagation algorithm. Furthermore, this dynamic approach to training is to be contrasted with a static approach, wherein a least square regression estimate for parameters is obtained in the limit of an infinitesimal time integration window. In a first test of static and dynamic approaches against ground truth data generated by the model, the former proves to be significantly faster and more accurate than the latter at recovering the parameters. When both calibration approaches are tested against DNS data, for which it is known that the model cannot achieve a perfect fit, the static approach yields a good prediction only for short times, while the dynamic approach results in physical and stable predictions over the entire integration window. After optimization of the dynamic approach for time step, spatial resolution, stability, and physics-based constraints, we obtain a 50% improvement of outcomes over those obtained from the existing, manually calibrated set of parameters, demonstrating the merits of this systematic and automated procedure.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Automated Detection of Qubit Structures on Quantum Chips

In this study, YOLO (You Only Look Once), a well-known object detection and image segmentation model, is used to detect qubits on a quantum chip. The model is trained by exposing it to various images of objects of interest and by setting various training parameters to allow the model to learn in unconventional conditions. We were able to provide the model with numerous images of qubits and various prominent features of note located on a qubit. We then created a simple GUI to display the model's live detections that would be integrated with the SEMs. This allowed us to create various functions such as auto-imagery of the qubits across a chip, auto-focusing, auto-contrasting, automatic staging, and directional corrections, enabling users to scan and analyze qubit surfaces in a fast and effective manner.

Perjuste, Ruth [Wellesley Coll.]↗
DOE OSTI2025

A Machine Learning Framework for Modeling Ensemble Properties of Atomically Disordered Materials

Atomic disorder can strongly influence material properties such as charge transport, optical response, and catalytic activity. However, efficiently modeling these disorder effects remains challenging for first-principles methods due to the cost of sampling large configurational spaces and computing complex physical quantities. Recent advances of machine learning techniques, particularly graph neural networks (GNNs), has enabled the efficient and accurate predictions of complex material properties, offering promising tools for studying disordered systems. In this work, we present a general machine-learning-assisted computational framework that integrates equivariant GNNs with Monte Carlo simulations to compute the thermodynamic and ensemble-averaged functional properties of disordered materials. Using the surface-termination-disordered MXene monolayer Ti 3 C 2 T 2–x as a representative system, we find that electrical conductivity exhibits an emergent peak near the order–disorder phase transition temperature due to the interplay between electron scattering and doping. In contrast, optical conductivity remains largely insensitive to local atomic disorder and reflects the global surface chemical composition. These results highlight the role of atomic disorder in affecting material properties and demonstrate the potential of our approach for statistically modeling disorder effects in a wide range of materials such as high-entropy alloys and spin liquids.

MXene↗
DOE OSTI2026

Stability Constants of Lanthanide-nitrate Complexes in Aqueous Solutions: A Theoretical Study

Calculating stability constants for lanthanide--nitrate complexes in aqueous solution is challenging due to the complex free-energy landscapes of the participating species. In this work, we compare cluster-continuum solvation and condensed-phase approaches using universal machine learning interatomic potentials (MLIPs) for determining the stability constants of lanthanide--nitrate complexes in aqueous solutions. Within the cluster--continuum solvation framework at the B3LYP level of theory, reactions involving lanthanide coordination numbers of both 8 and 9 are found to be relevant. After an empirical linear free-energy correction, the cluster--continuum results fall on the same order-of-magnitude scale as the experimental stability constants. By contrast, MACE-MP0 MLIP underestimates lanthanide hydration numbers and gives PMF-derived stability constants with large deviations from experiment, whereas MACE-MATPES-R2SCAN improves both hydration structure and the stability-constant scale but still does not quantitatively reproduce the detailed lanthanide trend. Across both approaches, nitrate binding is best viewed as a labile coordination motif rather than a fixed mono- or bidentate structure, with hydration-shell structure influencing which configurations are favored. Overall, the cluster--continuum calculations provide a practical semi-quantitative baseline for the experimental stability-constant scale, while the explicit-solvent MLIP benchmarks show clear progress from MACE-MP0 to MACE-MATPES-r2SCAN but also highlight the need for lanthanide-targeted training, fine-tuning, or improved long-range and polarization treatments to obtain predictive thermodynamics for complex aqueous lanthanide chemistry.

Dinpajooh, Mohammadhasan↗
DOE OSTI2026

Extended Rice–Thomson analysis and atomistic simulations revealing grain boundary effects on fracture in refractory high-entropy alloys

Significance This work serves to extend the fundamental ductile vs. brittle fracture theory, specifically the Rice–Thomson criterion, by introducing a grain boundary ahead of an initiating crack which propagates at an oblique angle to impinge the boundary. Atomistic fracture simulations on two refractory complex concentrated alloys, the brittle NbMoTaW and the ductile Nb 45 Ta 25 Ti 15 Hf 15 , demonstrate qualitative correspondence with the extended Rice–Thomson criterion and experimental observations. Abstract Understanding how grain boundaries mediate fracture remains a critical challenge in designing ductile, high-performance refractory alloys. Here, we extend the Rice–Thomson criterion to account for the angle between cracks and the impinging grain boundaries (GBs), capturing the competition between intergranular fracture and dislocation-mediated plasticity. Using machine learning interatomic potentials, we performed molecular statics simulations to probe fracture mechanisms in nanocrystalline NbMoTaW and Nb 45 Ta 25 Ti 15 Hf 15 , each with two different grain sizes, revealing trends consistent with experimental observations and the extended Rice model. Comparison with averaged R-curves for bulk samples demonstrates that GBs enhance ductility in Nb 45 Ta 25 Ti 15 Hf 15 in both grain sizes investigated. In contrast, GBs only locally improve fracture resistance in NbMoTaW when cracks are temporarily pinned at GBs inclined at high angles from the crack, but generally promote brittle intergranular fracture. These contrasting behaviors are attributed to differences in GB cohesion, reflecting clear alloying trends that align with ab-initio calculations and trends observed experimentally. Our results bridge classical fracture theory, atomistic simulations, and experimental observations, providing a comprehensive understanding of the fracture mechanisms in nanocrystalline refractory complex concentrated alloys.

36 MATERIALS SCIENCE↗
DOE OSTI2025

An analysis of physics limited dispatch of nuclear renewable integrated energy systems using deep reinforcement learning and dynamic modeling

Previous approaches to dispatching nuclear integrated energy systems (NIES) have focused on the profitability and flexibility of these systems to operate on energy grids with highly variable pricing. However, due to the complexity involved in modeling and designing these systems, there has been less emphasis on ensuring that these dispatch strategies are physically achievable. It is imperative to develop methods that allow the system to remain within the desired NIES operating conditions and perform this based on realistic limited forecasted information. This research employs next generation artificial intelligence, namely deep reinforcement learning (DRL), and a dynamic system model written in Modelica to find a safe and profitable dispatch strategy for a solar nuclear hybrid design. The DRL agent is shown to find a novel dispatch strategy that manages both power ramping and power levels while respecting operational limits. This DRL-based dispatch is compared to other dispatching strategies including an optimal design solution from mixed integer linear programming (MILP). It is found that incorporating the physics of such a tightly coupled NIES limits the profitability of the MILP-based dispatch strategy. As a result, the MILP solution overestimates the design’s generated revenue. In contrast, DRL significantly reduces the number of breaches of safe operational conditions during energy arbitrage while maintaining profitability. Furthermore, this work paves the way for a more detailed assessment of NIES profitability and could be used to aid operator decisions on future NIES projects.

14 - SOLAR ENERGY↗
DOE OSTI2025

Nature of molybdenum carbide surfaces for catalytic hydrogen dissociation using machine-learned potentials: an ensemble-averaged perspective

Molybdenum carbides with an electronic structure similar to noble metals have gained attention as a promising low-cost catalyst for biomass valorization and the hydrogen evolution reaction. However, our fundamental understanding of the catalyst surface and how different phases of these catalysts behave at varying reaction conditions is limited to ground state density functional theory calculations as ab initio molecular dynamics (AIMD) is computationally prohibitive at relevant length and time scales. Here, in this work, we train a multi-atomic cluster expansion (MACE) machine-learned interatomic potentials (MLIP) to study hydrogen dissociation and dynamics over Mo, δ-MoC, α-Mo 2 C, and β-Mo 2 C surfaces at varying temperatures and hydrogen partial pressures. Our simulations identify unique and different molecular and atomic hydrogen adsorption sites on different surfaces that do not depend on the temperature. At low hydrogen pressures, the surface coverage is monolayer, which transitions to two-layer adsorption at higher pressures. We find that atomic hydrogen diffusion and recombinations are preferred over molybdenum atom hollow sites, while the diffusion over carbon-terminated facets was negligible, signifying particularly strong C–H interactions. In contrast, molecular hydrogen adsorption occurs mostly atop Mo or the bridging sites. At a comparable hydrogen loading, β-Mo 2 C (001) is the most active surface for hydrogen dissociation reaction. This work provides insights into the dynamic nature of the hydrogen dissociation chemistry and the diversity of hydrogen adsorption sites on molybdenum carbides.

08 HYDROGEN↗
DOE OSTI2020

Remote Sensing-Informed Zonation for Understanding Snow, Plant and Soil Moisture Dynamics within a Mountain Ecosystem

In the headwater catchments of the Rocky Mountains, plant productivity and its dynamics are largely dependent upon water availability, which is influenced by changing snowmelt dynamics associated with climate change. Understanding and quantifying the interactions between snow, plants and soil moisture is challenging, since these interactions are highly heterogeneous in mountainous terrain, particularly as they are influenced by microtopography within a hillslope. Recent advances in satellite remote sensing have created an opportunity for monitoring snow and plant dynamics at high spatiotemporal resolutions that can capture microtopographic effects. In this study, we investigate the relationships among topography, snowmelt, soil moisture and plant dynamics in the East River watershed, Crested Butte, Colorado, based on a time series of 3-meter resolution PlanetScope normalized difference vegetation index (NDVI) images. To make use of a large volume of high-resolution time-lapse images (17 images total), we use unsupervised machine learning methods to reduce the dimensionality of the time lapse images by identifying spatial zones that have characteristic NDVI time series. We hypothesize that each zone represents a set of similar snowmelt and plant dynamics that differ from other identified zones and that these zones are associated with key topographic features, plant species and soil moisture. We compare different distance measures (Ward and complete linkage) to understand the effects of their influence on the zonation map. Results show that the identified zones are associated with particular microtopographic features; highly productive zones are associated with low slopes and high topographic wetness index, in contrast with zones of low productivity, which are associated with high slopes and low topographic wetness index. The zones also correspond to particular plant species distributions; higher forb coverage is associated with zones characterized by higher peak productivity combined with rapid senescence in low moisture conditions, while higher sagebrush coverage is associated with low productivity and similar senescence patterns between high and low moisture conditions. In addition, soil moisture probe and sensor data confirm that each zone has a unique soil moisture distribution. This cluster-based analysis can tractably analyze high-resolution time-lapse images to examine plant-soil-snow interactions, guide sampling and sensor placements and identify areas likely vulnerable to ecological change in the future.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Domain-wall-mediated polarization switching in ferroelectric AlScN: Strain relief and field-dependent dynamics

While scandium-doped aluminum nitride (AlScN) exhibits robust ferroelectricity and excellent thermal stability, its utility is limited by an exceptionally high coercive field (𝐸 𝑐 ) for polarization switching. Unraveling the atomistic switching dynamics is therefore critical for tailoring 𝐸 𝑐 . Here, in this study, we combine density functional theory and machine-learning molecular dynamics to elucidate the polarization switching mechanisms in AlScN over various Sc concentrations and applied electric fields. We find that excessive lattice strain strictly prohibits collective polarization switching, but the preexisting domain walls relieve strain and lead to a distinct switching dynamics—dictating a field-dependent switching mechanism. At low electric fields, switching occurs via gradual domain-wall propagation consistent with the Kolmogorov-Avrami-Ishibashi model. In contrast, high fields stimulate additional nucleation, driving a rapid, homogeneous reversal process described by the simultaneous nonlinear nucleation and growth model. These findings highlight the critical role of domain-wall dynamics and suggest domain engineering as a viable strategy to tailor coercive fields in AlScN and related ferroelectrics.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Evolving Architectures and Considerations to address Distributed Energy Resources and Non-Wired Alternatives

The electric grid is in the beginning stage of a transformation, driven by a combination of shutdowns of coal-fired plants, commissioning of new natural-gas plants, and tremendous growth in energy supply from renewables such as wind, and solar. As utilities navigate this transformation, their progress is supported by advances in Operational Technologies (OT), and Informational Technologies (IT), such as automation, smart inverters, cloud computing, mobile computing, machine learning, big data analytics, which have the potential to enable advanced capabilities more efficiently and at a lower cost. This white paper focuses on the architectural considerations that will allow the industry to transition in a planned manner. It introduces and formalizes two architectural constructs –the data bus and the control bus. The data bus is responsible for carrying all non-operational models and information necessary to drive utility decisions. In contrast, the control bus is responsible for carrying all operational data and control actions taken at the local level, centralized level, or other levels in-between. The paper reviews those architectural considerations, their requirements and how they will evolve. This paper intends to provide a context for vendors, utilities, and their service providers to review and understand the changes that are coming and get ready for them.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Exploring the formation of gold/silver nanoalloys with gas-phase synthesis and machine-learning assisted simulations

While nanoalloys are of paramount scientific and practical interest, the main processes leading to their formation are still poorly understood. Key structural features in the alloy systems, including the crystal phase, chemical ordering, and morphology, are challenging to control at the nanoscale, making it difficult to extend their use to industrial applications. Here, in this contribution, we focus on the gold/silver system that has two of the most prevalent noble metals and combine experiments with simulations to uncover the formation mechanisms at the atomic level. Nanoparticles were produced using a state-of-the-art inert-gas aggregation source and analyzed using transmission electron microscopy and energy-dispersive X-ray spectroscopy. Machine-learning-assisted molecular dynamics simulations were employed to model the crystallization process from liquid droplets to nanocrystals. Our study finds a preponderance of nanoparticles with five-fold symmetric morphology, including icosahedra and decahedra which is consistent with previous results on mono-metallic nanoparticles. However, we observed that gold atoms, rather than silver atoms, segregate at the surface of the obtained nanoparticles for all the considered alloy compositions. These segregation tendencies are in contrast to previous studies and have consequences on the crystallization dynamics and the subsequent crystal ordering. We finally showed that the underpinning of this surprising segregation dynamics is due to charge transfer and electrostatic interactions rather than surface energy considerations.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Pre-training Vision Models for the Classification of Alerts from Wide-field Time-domain Surveys

Modern wide-field time-domain surveys facilitate the study of transient, variable and moving phenomena by conducting image differencing and relaying alerts to their communities. Machine learning tools have been used on data from these surveys and their precursors for more than a decade, and convolutional neural networks (CNNs), which make predictions directly from input images, saw particularly broad adoption through the 2010s. Since then, continually rapid advances in computer vision have transformed the standard practices around using such models. It is now commonplace to use standardized architectures pre-trained on large corpora of everyday images (e.g., ImageNet). In contrast, time-domain astronomy studies still typically design custom CNN architectures and train them from scratch. Here, we explore the effects of adopting various pre-training regimens and standardized model architectures on the performance of alert classification. We find that the resulting models match or outperform a custom, specialized CNN like what is typically used for filtering alerts. Moreover, our results show that pre-training on galaxy images from Galaxy Zoo tends to yield better performance than pre-training on ImageNet or training from scratch. We observe that the design of standardized architectures are much better optimized than the custom CNN baseline, requiring significantly less time and memory for inference despite having more trainable parameters. On the eve of the Legacy Survey of Space and Time and other image-differencing surveys, these findings advocate for a paradigm shift in the creation of vision models for alerts, demonstrating that greater performance and efficiency, in time and in data, can be achieved by adopting the latest practices from the computer vision field.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Isotope effects in x-ray absorption spectra of liquid water

The isotope effects in x-ray absorption spectra of liquid water are studied by a many-body approach within electron-hole excitation theory. The molecular structures of both light and heavy water are modeled by path-integral molecular dynamics based on the advanced deep-learning technique. The neural network is trained on ab initio data obtained with SCAN density functional theory. The experimentally observed isotope effect in x-ray absorption spectra is reproduced semiquantitatively in theory. Compared to the spectrum in normal water, the blueshifted and less pronounced pre- and main-edge in heavy water reflect that the heavy water is more structured at short- and intermediate-range of the hydrogen-bond network. In contrast, we find the isotope effect on the spectrum is negligible at post-edge, which is consistent with the identical long-range ordering in both liquids as observed in the diffraction experiment.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2023

Fission with Exotic Nuclei (Abbreviated Report)

Nuclear fission is a key mechanism involved in the synthesis of heavy elements in the Cosmos and is the primary explanation for the stability of superheavy elements. Nevertheless, our knowledge of fission remains extremely fragmented. Most experiments have been conducted only on a tiny number of stable actinide nuclei and are often incomplete, leading to gaps in our basic understanding of the process. For many radioactive isotopes, basic fission data such as the charge or mass distribution of the fragments is unknown. These gaps cannot always be filled by simulation alone. Common fission models contain too many free parameters and lack predictive power. In contrast, the fundamental theory of fission under development at LLNL is much more predictive, but its current computational cost is too high to be used extensively for data evaluations. A unique window of opportunity to resolve these limitations has recently opened: the U.S. nuclear science community is ramping up major experimental programs at the Facility for Rare Isotope Beams (FRIB, the DOE flagship facility in low-energy nuclear science), and techniques from machine learning have shown great potential to simplify the use of a fundamental, quantum-mechanical theory of fission. This project has two components. On the experimental side, we acquired and deployed at the HIGS facility a new dual Frisch-Grid ionization chamber to measure correlated fragment-mass, kinetic energy, and angular distributions of fission fragments from induced fission. This new device was used to perform measurements of charge, mass and total kinetic energy of fission fragments in the photofission of 238 U and eight gamma-ray beam energies between 6.2 and 13 MeV, which allowed extracting high-precision independent yields for this reaction. The device was also used to perform measurements of the same quantities in the neutron-induced fission of 234 U with monoenergetic beams of energy between 5 and 8 MeV. In parallel, we collaborated with a team at Commissariat à l’énergie atomique et aux énergies alternatives (CEA) to perform a series of measurements of fission yields in inverse kinematics for the two isotopes of 236 U and 240 Pu. The experiment took place at the Grand Accélérateur National d’Ions Lourds in France in June 2023. The deployment of the VAMOS spectrometer with a new array called PISTA allowed determining the excitation energy of the fissioning system within 1 Mega-electronvolts. The second component of the project involved using deep neural networks to build fast and reliable emulators of our current fission models. In an invited paper published in Frontier in Physics, we showed that autoencoders could successfully compress nuclear wavefunctions in nuclear density functional theory. We achieved a dimensionality reduction of the order of two orders of magnitude while keeping the error in the total energy to less than 0.01%. In a second paper submitted to Physical Review Letters in June 2023 with our collaborators at CEA, we showed that variational autoencoders can learn the collective degrees of freedom driving the fission process.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

Unifying Combinatorial and Graphical Methods in Artificial Intelligence

Recently, a new graph Laplacian, called the inner product Laplacian, was introduced which generalizes many existing Laplacians, including the normalized and combinatorial Laplacian and their weighted variants. The key observation behind the inner product Laplacian is that by defining appropriate inner product spaces on the vertices and edges, the standard Laplacians can be recovered as Hodge Laplacians over the simplicial complex formed by the edges and vertices. These inner product spaces form a natural way to incorporate non-combinatorial information into the definition of a domain-specific Laplacian. In particular, in contrast to current domain-specific weighting schemes which rely solely on edge weights, information regarding the similarity of non-adjacent vertices and arbitrary pairs of edges can be effectively incorporated into the Laplacian. In order to illustrate this approach we consider the problem of calculating the potential energy of an atomistic configuration using Graph Neural Networks. In comparison with start-of-the-art approaches, such as SchNet, our approach replaces a learned (via auto-encoder) representation of the atom types with an inner product space on atoms based on scientific knowledge (e.g., electronegativity). We will illustrate how this approach captures key chemical properties of the molecules and compare the energy calculations with state-of-the-art neural network approaches. However, to compute the resulting Laplacian involves a mixture of sparse and dense matrix computation and yields a dense matrix as the basis for the graph convolution. This dense convolutional kernel necessitates moving away from the standard message passing framework for graph neural networks and increases the computational cost of applying the kernel. In order to mitigate these costs we investigate means of leveraging the mixed sparse and dense computations to reduce the overall computational cost and how these approaches can be automatically transferred to energy efficient hardware (e.g., field programmable gate arrays (FPGAs)).

97 MATHEMATICS AND COMPUTING↗