Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep learning method”

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 1,765 records · Page 98

Probabilistic Prediction of Geomagnetic Storms and the K p index

Geomagnetic activity is often described using summary indices to summarize the likelihood of space weather impacts, as well as when parameterizing space weather models. The geomagnetic index K p in particular, is widely used for these purposes. Current state-of-the-art forecast models provide deterministic K p predictions using a variety of methods – including empirically-derived functions, physics-based models, and neural networks – but do not provide uncertainty estimates associated with the forecast. This paper provides a sample methodology to generate a 3-hour-ahead K p prediction with uncertainty bounds and from this provide a probabilistic geomagnetic storm forecast. Specifically, we have used a two-layered architecture to separately predict storm (K p ≥ 5 – ) and non-storm cases. As solar wind-driven models are limited in their ability to predict the onset of transient-driven activity we also introduce a model variant using solar X-ray flux to assess whether simple models including proxies for solar activity can improve the predictions of geomagnetic storm activity with lead times longer than the L1-to-Earth propagation time. By comparing the performance of these models we show that including operationally-available information about solar irradiance enhances the ability of predictive models to capture the onset of geomagnetic storms and that this can be achieved while also enabling probabilistic forecasts.

79 ASTRONOMY AND ASTROPHYSICS↗

Algorithm-Based Fault Tolerance for Convolutional Neural Networks

Convolutional neural networks (CNNs) are becoming more and more important for solving challenging and critical problems in many fields. CNN inference applications have been deployed in safety-critical systems, which may suffer from soft errors caused by high-energy particles, high temperature, or abnormal voltage. Of critical importance is ensuring the stability of the CNN inference process against soft errors. Traditional fault tolerance methods are not suitable for CNN inference because error-correcting code is unable to protect computational components, instruction duplication techniques incur high overhead, and existing algorithm-based fault tolerance (ABFT) techniques cannot protect all convolution implementations. In this paper, we focus on how to protect the CNN inference process against soft errors as efficiently as possible, with the following three contributions. (1) We propose several systematic ABFT schemes based on checksum techniques and analyze their fault protection ability and runtime thoroughly. Unlike traditional ABFT based on matrix-matrix multiplication, our schemes support any convolution implementations. (2) We design a novel workflow integrating all the proposed schemes to obtain a high detection/correction ability with limited total runtime overhead. (3) We perform our evaluation using ImageNet with well-known CNN models including AlexNet, VGG-19, ResNet-18, and YOLOv2. Here, experimental results demonstrate that our implementation can handle soft errors with very limited runtime overhead (4%~8% in both error-free and error-injected situations).

97 MATHEMATICS AND COMPUTING↗

A data-driven method for modelling dissipation rates in stratified turbulence

We present a deep probabilistic convolutional neural network (PCNN) model for predicting local values of small-scale mixing properties in stratified turbulent flows, namely the dissipation rates of turbulent kinetic energy and density variance, $\varepsilon$ and $\chi$ . Inputs to the PCNN are vertical columns of velocity and density gradients, motivated by data typically available from microstructure profilers in the ocean. The architecture is designed to enable the model to capture several characteristic features of stratified turbulence, in particular the dependence of small-scale isotropy on the buoyancy Reynolds number $Re_b:=\varepsilon /(\nu N^2)$ , where $\nu$ is the kinematic viscosity and $N$ is the background buoyancy frequency, the correlation between suitably locally averaged density gradients and turbulence intensity and the importance of capturing the tails of the probability distribution functions of values of dissipation. Empirically modified versions of commonly used isotropic models for $\varepsilon$ and $\chi$ that depend only on vertical derivatives of density and velocity are proposed based on the asymptotic regimes $Re_b\ll 1$ and $Re_b\gg 1$ , and serve as an instructive benchmark for comparison with the data-driven approach. When trained and tested on a simulation of stratified decaying turbulence which accesses a range of turbulent regimes (associated with differing values of $Re_b$ ), the PCNN outperforms assumptions of isotropy significantly as $Re_b$ decreases, and additionally demonstrates improvements over the fitted empirical models. A differential sensitivity analysis of the PCNN facilitates a comparison with the theoretical models and provides a physical interpretation of the features enabling it to make improved predictions.

42 ENGINEERING↗

Achieving 50% Energy Savings in Chicago Homes: A Case Study for Advancing Equity and Climate Goals

Since 2019, the National Renewable Energy Laboratory (NREL) and Elevate have collaborated to identify pathways to deep energy retrofits in Chicago's housing stock, document equity implications and co-benefits of this transition, and validate the findings by implementing retrofits in real Chicago homes. This document summarizes our analysis process to model advanced retrofit packages that lead to greater than 50% energy savings in Chicago homes. Based on these findings, we have also developed a roadmap with the City to guide implementation, and are deploying the recommended retrofit packages in real Chicago homes to realize these energy savings. This work was developed in collaboration with two key stakeholders - the City of Chicago and Commonwealth Edison (ComEd) - and funded by the U.S. Department of Energy (DOE). NREL's Residential Buildings team maintains the best-in-class ResStock™ energy model of the U.S. residential building stock. For this work, we calibrated ResStock to Chicago's unique local housing stock to accurately simulate energy use in Chicago homes both for current conditions and with various retrofit scenarios. We simulated a wide range of potential building retrofits covering all aspects of residential energy use and then grouped these into packages based on energy and utility bill savings. ResStock can model diverse building types and housing characteristics, so we're able to observe the range of outcomes that might occur when these upgrades are deployed across the entire housing stock. We can then estimate potential energy savings from an advanced retrofit program on Chicago's housing stock by comparing the modeled energy use before versus after a retrofit. This novel version of ResStock, calibrated to Chicago with data from Elevate, can help City officials, ComEd, and other partners plan for community-scale decarbonization via residential retrofits. Specifically, this work contributes the following project goals: Develop a building retrofit prioritization strategy for Chicago single-family and 2- to 4-unit buildings; Identify neighborhoods and home types that have the highest potential for savings from electrification; and Assess the impact of advanced building retrofits on energy use, utility bills, and CO2 emissions at the city and building level. Although this study is specific to Chicago, its methods and learnings are applicable across the United States. These findings are especially notable for heat pumps and electrification retrofits in cold climates.

building energy modeling↗

Efficient mapping between void shapes and stress fields using Deep Convolutional Neural Networks with sparse data

Establishing fast and accurate structure-to-property relationships is an important component in the design and discovery of advanced materials. Physics-based simulation models like the finite element method (FEM) are often used to predict deformation, stress, and strain fields as a function of material microstructure in material and structural systems. Such models may be computationally expensive and time intensive if the underlying physics of the system is complex. This limits their application to solve inverse design problems and identify structures that maximize performance. In such scenarios, surrogate models are employed to make the forward mapping computationally efficient to evaluate. However, the high dimensionality of the input microstructure and the output field of interest often renders such surrogate models inefficient, especially when dealing with sparse data. Deep convolutional neural network (CNN) based surrogate models have shown great promise in handling such high-dimensional problems. In this paper, a single ellipsoidal void structure under a uniaxial tensile load represented by a linear elastic, high-dimensional and expensive-to-query, FEM model. We consider two deep CNN architectures, a modified convolutional autoencoder framework with a fully connected bottleneck and a UNet CNN, and compare their accuracy in predicting the von Mises stress field for any given input void shape in the FEM model. Additionally, a sensitivity analysis study is performed using the two approaches, where the variation in the prediction accuracy on unseen test data is studied through numerical experiments by varying the number of training samples from 20 to 100.

surrogate modeling; convolutional neural networks;↗

Topics in Theoretical Physics

This award supported a broadly based research effort in theoretical particle physics, including research aimed at uncovering the laws of nature at short (subatomic) and long (cosmological) distances. These theoretical developments apply to experiments in laboratories such as CERN, the facility that operates the Large Hadron Collider outside Geneva, as well as to cosmological investigations done using telescopes and satellites. The results reported here apply to physics beyond the so-called Standard Model of particle physics; physics of high energy collisions such as those observed at the Large Hadron Collider; theoretical and mathematical tools and frameworks for describing the laws of nature at short distances; cosmology and astrophysics; and analytic and computational methods to solve theories of short distance physics. Some specific research accomplishments include + Theories of the electroweak interactions, the forces that give rise to many forms of radioactive decay; + Physics of the recently discovered Higgs boson. + Models and phenomenology of dark matter, the mysterious component of the universe, that has so far been detected only by its gravitational effects. + High energy particles in astrophysics and cosmology. + Algorithmic research and Computational methods for physics of and beyond the Standard Model. + Theory and applications of relativity and its possible limitations. + Topological effects in field theory and cosmology. + Conformally invariant systems and AdS/CFT. This award also supported significant training of students and postdoctoral fellows to lead the research effort in particle theory for the coming decades. These students and fellows worked closely with other members of the group as well as theoretical and experimental colleagues throughout the physics community. Many of the research projects funded by this grant arose in response to recently obtained experimental results in the areas of particle physics and cosmology. We describe a few of these below. Relativity is founded on a symmetry property of nature called "Lorentz Invariance". Like all symmetry properties, it is essential to determine precisely how symmetric nature actually is; that is, do the laws of nature fully respect the symmetry or is there room for tiny symmetry violating effects? An important consequence of Lorentz invariance is the existence of a universal limiting velocity for all physical particles. Light travels at this limiting velocity so it is frequently referred to as simply "the speed of light", but relativity requires that ALL particles travel more slowly than this speed. Once the Higgs particle was discovered in 2012 a natural question was whether or not this particle's speed was consistent with relativity. Although the speed of the Higgs particle is not measurable directly, Cohen has shown that, if the maximal speed of the Higgs particle was not precisely the same as the speed of light, then the Higgs would have some unusual properties. In some cases the Higgs would be unstable to some unusual decay modes; in other cases the interactions of the Higgs with other particles would change the properties of these other particles in ways that could be observed in so-called cosmic rays, very energetic particles (such as photons, protons and other atomic nuclei) coming from space. Once these particles hit the upper atmosphere they produce a "shower" of particles that can be seen by ground-based instruments. If the Higgs has a maximal speed that differs even a tiny bit from the speed of light these showers would look quite different from what is observed. In this way Cohen was able to establish that the Higgs travels with a maximal speed that cannot differ from the speed of light by more than one part in a thousand-trillion. This is by far the most precisely determined property of the Higgs particle. Cohen and Schmaltz reviewed evidence from the Large Hadron Collider (LHC), a particle physics experiment operating at the CERN laboratory near Geneva, for a new particle sometimes called a W'. This evidence included certain unexpected by-products in collisions of protons at very high energy. While the evidence was not significant enough to claim a discovery, it was sufficiently intriguing that many particle theorists worked to construct explanations for this signal. Cohen and Schmaltz were able to determine that such explanations are highly constrained by previous experiments involving collisions of very energetic particles. Nevertheless they were able to construct a theory that adequately explains the LHC data and remain consistent with prior experiments. Their explanation predicts the existence of yet another new particle, called a Z', with a mass slightly greater than that of the W'. This additional particle, if it exists, should be seen as more data is collected from the LHC. Amusingly, there is one collision by-product that has already been seen by the CMS experiment at the LHC that supports the existence of this new particle; however, it is not unlikely that this single event is a so-called "background" event, that is a somewhat atypical by-product of a conventional Standard Model process. This theory for the anomalous LHC data will either be confirmed or excluded with further data-taking at the LHC. The ratio of the number of electrons produced in bottom quark decays over the number of muons produced has been measured at the LHC. This ratio is interesting because it can be predicted very precisely from a basic property of the Standard Model: lepton universality. If lepton universality is correct, the ratio of electrons to muons is predicted to be equal to 1. The first measurements of this ratio find a value different from 1 with a statistical significance of about 3 standard deviations. Schmaltz and collaborator proposed a new extension of the Standard Model which can explain the new data. In addition, Schmaltz and collaborators proposed several new measurements of ratios of decay rates which can confirm or rule out the surprising results from the earlier LHC data. The most recent and precise measurements of the cosmic microwave background from the Planck satellite, from a combination of measurements of the dark matter distribution in the universe, and from a measurement of the expansion rate of the universe today show some disagreement when interpreted in terms of the so-called LambdaCDM model. Schmaltz and collaborators proposed an alternative model to LambdaCDM in which the usual cold dark matter is replaced by a new ``dark sector". This sector consists of a cold dark matter particle which interacts with a newly postulated dark radiation component of the universe. The dark radiation can help explain the discrepancy in measurements of the expansion rate, and the dark matter interactions subtly modify the clumping of dark matter at large scales, thus potentially explaining both kinds of tensions in the data. In two publications Schmaltz described the new model and then performed a precision comparison of the predictions of the model with all currently available cosmological data. The results favor the new model at the level of three standard deviations with current data. Quantum Field Theory (QFT) is the language we use to describe quantum systems which are consistent with Einstein’s theory of Special Relativity. In particular, the requirement of Einstein’s theory that signals not travel faster than the speed of light constrains the types of interactions which particles can engage in. One consequence of relativity is that these interactions cannot preserve particle number. The stronger the interactions, the more severe the particle number violation in a given Relativistic QFT. When particle number violation is strong, it becomes very difficult to adequately parameterize the quantum wave function (which characterizes the state of a quantum system). For example, though we can formulate the QFT which describes the strong force as a set of interactions between quark and gluon particles, we have no clear idea how to express the proton state in terms of these quarks and gluons. This is because the proton, though a bound state of quarks and gluons, is not a state of a fixed number of particles due to strong interactions. Yet, understanding the proton state is very important in order to theoretically predict the reaction rates observed at the LHC in Geneva, which is a proton-proton collider. Katz has formulated a new approach to QFT, which among other things offers a way to adequately approximate the quantum wave function of a bound state at strong coupling. The approximation scheme is related to the fact that any sensible QFT (including that of the strong interactions) is at short distances approximately self-similar upon rescaling of space and time. It turns out that keeping track of the response upon this rescaling is important in efficiently parameterizing the state. Katz and collaborators have used this observation to approximate the state of the proton in toy versions of the strong force. In the late 60s Sheldon Glashow, Abdus Salam and Steven Weinberg (1979 Nobel Prize awardees) proposed a theory unifying weak and electromagnetic interaction which assumed the existence of new particles, the W and Z bosons. The W and Z bosons were eventually detected in high-energy collision in a particle accelerator at CERN, and the recent discovery of the Higgs meson at the Large Hadron Collider (LHC), always at CERN, completed the picture. However, deep theoretical considerations indicate that the theory by Glashow, Weinberg and Salam, often referred to as "the standard model" cannot be the whole story: the existence of new particles and new interactions at yet higher energies is widely anticipated. The experiments at the LHC are looking for these, while theorists, like Brower, Rebbi and collaborators, are investigating models for these new interactions. Working in a large national collaboration with access to the most powerful DOE computers Brower, Rebbi and colleagues have been using calculational techniques, similar to those successfully employed for many years to investigate the interactions among quarks in nucleons, to study theories that can describe the expected "beyond the standard model" (BSM) interactions. Their results, which include also a model for dark matter, have been published in several refereed papers in prestigious journals. Various ideas in topologically interesting field theories predict hypothetical objects such as fractional charges and Majorana excitations. However, such fascinating objects have not been seen in particle physics. Nevertheless, these objects demonstrate possible phenomena that quantum field theory can support. Pi used condensed matter physics as a laboratory to study possible realizations and observable effects of these objects predicted by quantum field theory. In recent times there has developed considerable interest among condensed matter field theorists in precisely the same geometrical and topological structures, which were first discovered in particle physics field theories. From particle physicists' point of view, this is an interesting development, since condensed matter provides an arena in which one can concretely realize particle physics ideas. Moreover, particle physicists can learn new ideas from condensed matter physics. Higgs phenomenon is precisely an important particle physics realization of condensed matter ideas. In contrast to the small distance characterizing condensed matter systems, field theory also describes large distance physics characterizing cosmology. Pi worked on various geometrical effects in the standard theory of cosmology, viz general relativity.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Scaling and merging time-resolved pink-beam diffraction with variational inference

Time-resolved x-ray crystallography (TR-X) at synchrotrons and free electron lasers is a promising technique for recording dynamics of molecules at atomic resolution. While experimental methods for TR-X have proliferated and matured, data analysis is often difficult. Extracting small, time-dependent changes in signal is frequently a bottleneck for practitioners. Recent work demonstrated this challenge can be addressed when merging redundant observations by a statistical technique known as variational inference (VI). However, the variational approach to time-resolved data analysis requires identification of successful hyperparameters in order to optimally extract signal. In this case study, we present a successful application of VI to time-resolved changes in an enzyme, DJ-1, upon mixing with a substrate molecule, methylglyoxal. We present a strategy to extract high signal-to-noise changes in electron density from these data. Furthermore, we conduct an ablation study, in which we systematically remove one hyperparameter at a time to demonstrate the impact of each hyperparameter choice on the success of our model. We expect this case study will serve as a practical example for how others may deploy VI in order to analyze their time-resolved diffraction data.

47 OTHER INSTRUMENTATION↗

Fast meta-solvers for 3D complex-shape scatterers using neural operators trained on a non-scattering problem

Three-dimensional target identification using scattering techniques requires high accuracy solutions and very fast computations for real-time predictions in some critical applications. We first train a deep neural operator (DeepONet) to solve wave propagation problems described by the Helmholtz equation in a domain without scatterers but at different wavenumbers and with a complex absorbing boundary condition. We then design two classes of fast meta-solvers by combining DeepONet with either relaxation methods, such as Jacobi and Gauss-Seidel, or with Krylov methods, such as GMRES and BiCGStab, using the trunk basis of DeepONet as a coarse-scale preconditioner. We leverage the spectral bias of neural networks to account for the lower part of the spectrum in the error distribution while the upper part is handled inexpensively using relaxation methods or fine-scale preconditioners. The meta-solvers are then applied to solve scattering problems with different shape of scatterers, at no extra training cost. We first demonstrate that the resulting meta-solvers are shape-agnostic, fast, and robust, whereas the standard standalone solvers may even fail to converge without the DeepONet. We then apply both classes of meta-solvers to scattering from a submarine, a complex three-dimensional problem. We achieve very fast solutions, especially with the DeepONet-Krylov methods, which require orders of magnitude fewer iterations than any of the standalone solvers.

97 MATHEMATICS AND COMPUTING↗

Large-momentum effective theory

Since the parton model was introduced by Feynman more than 50 years ago, much has been learned about the partonic structure of the proton through a large body of high-energy experimental data and dedicated global fits. However, limited progress has been made in calculating partonic observables such as the parton distribution function (PDFs) from the fundamental theory of strong interactions, quantum chromodynamics (QCD). Recently some advocated for a formalism, large-momentum effective theory (LaMET), through which one can extract parton physics from the properties of the proton traveling at a moderate boost factor such as γ ~ 2 – 5 . The key observation behind this approach is that Lorentz symmetry allows the standard formalism of partons in terms of light-front operators to be replaced by an equivalent one with large-momentum states and time-independent operators of a universality class. With LaMET, the PDFs, generalized PDFs or generalized parton distributions, transverse-momentum-dependent PDFs, and light-front wave functions can all be extracted in principle from lattice simulations of QCD (or other nonperturbative methods) through standard effective field theory matching and running. Future lattice QCD calculations with exascale computational facilities could help one to understand the experimental data related to the hadronic structure, including those from the upcoming electron-ion colliders dedicated to exploring the partonic landscape of the proton. In this work, the progress made in the past few years in the development of the LaMET formalism and its applications is reviewed, with an emphasis on a demonstration of its effectiveness from initial lattice QCD simulations.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Baseflow Identification via Explainable AI With Kolmogorov‐Arnold Networks

Abstract Hydrological models often involve constitutive laws that may not be optimal in every application. We propose to replace such laws with the Kolmogorov‐Arnold networks (KANs), a class of neural networks designed to identify symbolic expressions. We demonstrate KAN's potential on the problem of baseflow identification, a notoriously challenging task plagued by significant uncertainty. KAN‐derived functional dependencies of the baseflow components on the aridity index outperform their original counterparts; they demonstrate that water availability, rather than potential evapotranspiration, drives baseflow by constraining actual evapotranspiration under arid conditions. On a test set, they increase the Nash‐Sutcliffe efficiency (NSE) by 65%, decrease the root mean squared error by 29%, and increase the Kling‐Gupta efficiency by 34%. This superior performance is achieved while reducing the number of fitting parameters from three to two. Next, we use data from 378 catchments across the continental United States to refine the water‐balance equation at the mean‐annual scale. The KAN‐derived equations based on the refined water balance outperform both the current aridity index model, with up to a 105% increase in NSE, and the KAN‐derived equations based on the original water balance. While the performance of our model and tree‐based machine learning methods is similar, KANs offer the advantage of simplicity and transparency and require no specific software or computational tools. This case study focuses on the aridity index formulation, but the approach is flexible and transferable to other hydrological processes. Plain Language Summary Equations used in hydrologic model are often suboptimal, resulting in reduced prediction accuracy and efficiency. We implemented Kolmogorov‐Arnold networks (KAN), a machine learning algorithm for deriving symbolic formulations, to estimate groundwater recharge and showed that it outperforms an existing state‐of‐the‐art semi‐empirical formulation. In hydrology, Nash‐Sutcliffe efficiency (NSE), root mean squared error (RMSE), and Kling‐Gupta efficiency (KGE) are commonly used to evaluate model performance. Higher NSE and KGE values indicate better performance, while lower RMSE values are preferable. Our results show that NSE increased by 71%, RMSE decreased by 32%, and KGE improved by 25%. In addition, KAN identifies an optimal functional form and can be used to derive new analytical formulas using the prior knowledge. The KAN‐inspired equation outperformed the original formulation and reduced the fitting parameters. Furthermore, we refined the water‐balance equation at the mean‐annual scale and showed that, based on the new water‐balance equation, KAN can derive new formulations that are superior to the original aridity index formulations (up to 105% increase in NSE) and KAN‐derived equations based on the original water balance. These findings highlight the significant potential of KAN to advance the scientific understanding of a wide range of hydrologic processes. Key Points Kolmogorov‐Arnold networks (KANs) enhance interpretability of machine‐learned hydrological models KAN‐derived symbolic formulations outperform state‐of‐the‐art semi‐empirical aridity indices KAN‐identified functional form yields an analytical index with fewer fitting parameters and improved performance

baseflow↗

Evaluating the Potential and Challenges of an Uncertainty Quantification Method for Long Short–Term Memory Models for Soil Moisture Predictions

Recently, recurrent deep networks have shown promise to harness newly available satellite–sensed data for long–term soil moisture projections. However, to be useful in forecasting, deep networks must also provide uncertainty estimates. Here we evaluated Monte Carlo dropout with an input–dependent data noise term (MCD+N), an efficient uncertainty estimation framework originally developed in computer vision, for hydrologic time series predictions. MCD+N simultaneously estimates a heteroscedastic input–dependent data noise term (a trained error model attributable to observational noise) and a network weight uncertainty term (attributable to insufficiently constrained model parameters). Although MCD+N has appealing features, many heuristic approximations were employed during its derivation, and rigorous evaluations and evidence of its asserted capability to detect dissimilarity were lacking. To address this, we provided an in–depth evaluation of the scheme's potential and limitations. We showed that for reproducing soil moisture dynamics recorded by the Soil Moisture Active Passive (SMAP) mission, MCD+N indeed gave a good estimate of predictive error, provided that we tuned a hyperparameter and used a representative training data set. The input–dependent term responded strongly to observational noise, while the model term clearly acted as a detector for physiographic dissimilarity from the training data, behaving as intended. However, when the training and test data were characteristically different, the input–dependent term could be misled, undermining its reliability. Additionally, due to the data–driven nature of the model, data noise also influences network weight uncertainty, and therefore the two uncertainty terms are correlated. Altogether, this approach has promise, but care is needed to interpret the results.

54 ENVIRONMENTAL SCIENCES↗

Toward a Machine Learning Approach to Interpreting X-ray Spectra of Trace Impurities by Converting XANES to EXAFS

The fact that the photoabsorption spectrum of a material contains information about the atomic structure, commonly understood in terms of multiple scattering theory, is the basis of the popular extended X-ray absorption spectroscopy (EXAFS) technique. How much of the same structural information is present in other complementary spectroscopic signals is not obvious. Here we use a machine learning approach to demonstrate that within theoretical models that accurately predict the EXAFS signal, the extended near-edge region does indeed contain the EXAFS-accessible structural information. We do this by exhibiting deep operator neural networks (DeepONets) that have learned the relationship between the extended and near edge portions of the X-ray absorption spectrum to predict the former from the latter. We find that we can accurately predict the EXAFS spectrum between 6 and 14 Å –1 from the first 6 Å –1 (≈100 eV) of the absorption spectrum of Cu 2 + substitutional defects in the Fe 3+ mineral hematite (α-Fe 2 O 3 ). This surprising finding implies that theoretical analyses of X-ray absorption spectra could be implemented that extract the same conclusions as high-quality EXAFS studies from spectra collected over a much smaller range of photon energies. This relaxes a host of experimental limitations related to the X-ray source and measurement sample, including collection time, minimum dopant concentration, source brilliance, and energy range. We describe the theoretical data sets and DeepONet construction and show that the resulting DeepONets produce EXAFS that recovers linear combination fits to experimental data with accuracy approaching the original ab initio calculations. We discuss the implications of our findings for minor constituent characterization and for understanding the information content of spectroscopic data more broadly, including how this approach might be applied to measured experimental spectra. In conclusion, to encourage similar efforts, the simulated X-ray spectra, machine learning, and fitting code are publicly available.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Reinforcement Learning-Based Oscillation Dampening: Scaling Up Single-Agent Reinforcement Learning Algorithms to a 100-Autonomous-Vehicle Highway Field Operational Test

In this article, we explore the technical details of the reinforcement learning (RL) algorithms that were deployed in the largest field test of automated vehicles designed to smooth traffic flow in history as of 2023, uncovering the challenges and breakthroughs that come with developing RL controllers for automated vehicles. We delve into the fundamental concepts behind RL algorithms and their application in the context of self-driving cars, discussing the developmental process from simulation to deployment in detail, from designing simulators to reward function shaping. We present the results in both simulation and deployment, discussing the flow-smoothing benefits of the RL controller. From understanding the basics of Markov decision processes to exploring advanced techniques such as deep RL, our article offers a comprehensive overview and deep dive of the theoretical foundations and practical implementations driving this rapidly evolving field. We also showcase real-world case studies and alternative research projects that highlight the impact of RL controllers in revolutionizing autonomous driving. From tackling complex urban environments to dealing with unpredictable traffic scenarios, these intelligent controllers are pushing the boundaries of what automated vehicles can achieve. Furthermore, we examine the safety considerations and hardware-focused technical details surrounding deployment of RL controllers into automated vehicles. As these algorithms learn and evolve through interactions with the environment, ensuring their behavior aligns with safety standards becomes crucial. Here, we explore the methodologies and frameworks being developed to address these challenges, emphasizing the importance of building reliable control systems for automated vehicles.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Solar Transposition Modeling via Deep Neural Networks With Sky Images

This article presents a notable advance toward the development of a new method of increasing the single-axis tracking photovoltaic (PV) system power output by improving the determination and near-term prediction of the optimum module tilt angle. The tilt angle of the plane receiving the greatest total irradiance changes with Sun position and atmospheric conditions including cloud formation and movement, aerosols, and particulate loading, as well as varying albedo within a module's field of view. In this article, we present a multi-input convolutional neural network that can create a profile of plane-of-array irradiance versus surface tilt angle over a full 180° arc from horizon to horizon. As input, the neural network uses the calculated solar position and clear-sky irradiance values, along with sky images. The target irradiance values are provided by the multiplanar irradiance sensor (MPIS). In order to account for varying irradiance conditions, the MPIS signal is normalized by the theoretical clear-sky global horizontal irradiance. Using this information, the neural network outputs an N -dimensional vector, where N is the number of points to approximate the MPIS curve via Fourier resampling. The output vector of the model is smoothed with a Gaussian kernel to account for error in the downsamping and subsequent upsampling steps, as well as to smooth the unconstrained output of the model. Furthermore, these profiles may be used to perform near-term prediction of angular irradiance, which can then inform the movement of a PV tracker.

42 ENGINEERING↗

Cu–Ni Oxidation Mechanism Unveiled: A Machine Learning-Accelerated First-Principles and in Situ TEM Study

Here, the development of accurate methods for determining how alloy surfaces spontaneously restructure under reactive and corrosive environments is a key, long-standing, grand challenge in materials science. Using machine learning-accelerated density functional theory and rare-event methods, in conjunction with in situ environmental transmission electron microscopy (ETEM), we examine the interplay between surface reconstructions and preferential segregation tendencies of CuNi(100) surfaces under oxidation conditions. Our modeling approach predicts that oxygen-induced Ni segregation in CuNi alloys favors Cu(100)-O c(2 × 2) reconstruction and destabilizes the Cu(100)-O (2√2 × √2)R45° missing row reconstruction (MRR). In situ ETEM experiments validate these predictions and show Ni segregation followed by NiO nucleation and growth in regions without MRR, with secondary nucleation and growth of Cu 2 O in MRR regions. Our approach based on combining disparate computational components and in situ ETEM provides a holistic description of the oxidation mechanism in CuNi, which applies to other alloy systems.

36 MATERIALS SCIENCE↗

Electromagnetic shower reconstruction and energy validation with Michel electrons and π 0 samples for the deep-learning-based analyses in MicroBooNE

This article presents the reconstruction of the electromagnetic activity from electrons and photons (showers) used in the MicroBooNE deep learning-based low energy electron search. The reconstruction algorithm uses a combination of traditional and deep learning-based techniques to estimate shower energies. We validate these predictions using two v μ -sourced data samples: charged/neutral current interactions with final state neutral pions and charged current interactions in which the muon stops and decays within the detector producing a Michel electron. Both the neutral pion sample and Michel electron sample demonstrate agreement between data and simulation. Finally, the absolute shower energy scale is shown to be consistent with the relevant physical constant of each sample: the neutral pion mass peak and the Michel energy cutoff.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Operator learning for predicting multiscale bubble growth dynamics

We report simulating and predicting multiscale problems that couple multiple physics and dynamics across many orders of spatiotemporal scales is a great challenge that has not been investigated systematically by deep neural networks (DNNs). Herein, we develop a framework based on operator regression, the so-called deep operator network (DeepONet), with the long-term objective to simplify multiscale modeling by avoiding the fragile and time-consuming “hand-shaking” interface algorithms for stitching together heterogeneous descriptions of multiscale phenomena. To this end, as a first step, we investigate if a DeepONet can learn the dynamics of different scale regimes, one at the deterministic macroscale and the other at the stochastic microscale regime with inherent thermal fluctuations. Specifically, we test the effectiveness and accuracy of the DeepONet in predicting multirate bubble growth dynamics, which is described by a Rayleigh–Plesset (R–P) equation at the macroscale and modeled as a stochastic nucleation and cavitation process at the microscale by dissipative particle dynamics (DPD). First, we generate data using the R–P equation for multirate bubble growth dynamics caused by randomly time-varying liquid pressures drawn from Gaussian random fields (GRFs). Our results show that properly trained DeepONets can accurately predict the macroscale bubble growth dynamics and can outperform long short-term memory networks. We also demonstrate that the DeepONet can extrapolate accurately outside the input distribution using only very few new measurements. Subsequently, we train the DeepONet with DPD data corresponding to stochastic bubble growth dynamics. Although the DPD data are noisy and we only collect sparse data points on the trajectories, the trained DeepONet model is able to predict accurately the mean bubble dynamics for time-varying GRF pressures. Taken together, our findings demonstrate that DeepONets can be employed to unify the macroscale and microscale models of the multirate bubble growth problem, hence providing new insight into the role of operator regression via DNNs in tackling realistic multiscale problems and in simplifying modeling with heterogeneous descriptions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

AICCA: AI-Driven Cloud Classification Atlas

Clouds play an important role in the Earth’s energy budget, and their behavior is one of the largest uncertainties in future climate projections. Satellite observations should help in understanding cloud responses, but decades and petabytes of multispectral cloud imagery have to date received only limited use. This study describes a new analysis approach that reduces the dimensionality of satellite cloud observations by grouping them via a novel automated, unsupervised cloud classification technique based on a convolutional autoencoder, an artificial intelligence (AI) method good at identifying patterns in spatial data. Our technique combines a rotation-invariant autoencoder and hierarchical agglomerative clustering to generate cloud clusters that capture meaningful distinctions among cloud textures, using only raw multispectral imagery as input. Cloud classes are therefore defined based on spectral properties and spatial textures without reliance on location, time/season, derived physical properties, or pre-designated class definitions. We use this approach to generate a unique new cloud dataset, the AI-driven cloud classification atlas (AICCA), which clusters 22 years of ocean images from the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Aqua and Terra instruments—198 million patches, each roughly 100 km × 100 km (128 × 128 pixels)—into 42 AI-generated cloud classes, a number determined via a newly-developed stability protocol that we use to maximize richness of information while ensuring stable groupings of patches. AICCA thereby translates 801 TB of satellite images into 54.2 GB of class labels and cloud top and optical properties, a reduction by a factor of 15,000. The 42 AICCA classes produce meaningful spatio-temporal and physical distinctions and capture a greater variety of cloud types than do the nine International Satellite Cloud Climatology Project (ISCCP) categories—for example, multiple textures in the stratocumulus decks along the West coasts of North and South America. We conclude that our methodology has explanatory power, capturing regionally unique cloud classes and providing rich but tractable information for global analysis. AICCA delivers the information from multi-spectral images in a compact form, enables data-driven diagnosis of patterns of cloud organization, provides insight into cloud evolution on timescales of hours to decades, and helps democratize climate research by facilitating access to core data.

97 MATHEMATICS AND COMPUTING↗