Search NASA⌕ Search

SEARCH · Search NASA

Results for “Explainable machine learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 379 records · Page 21

Lorentz group equivariant autoencoders

Abstract There has been significant work recently in developing machine learning (ML) models in high energy physics (HEP) for tasks such as classification, simulation, and anomaly detection. Often these models are adapted from those designed for datasets in computer vision or natural language processing, which lack inductive biases suited to HEP data, such as equivariance to its inherent symmetries. Such biases have been shown to make models more performant and interpretable, and reduce the amount of training data needed. To that end, we develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant with respect to the proper, orthochronous Lorentz group $$\textrm{SO}^+(3,1)$$ SO + ( 3 , 1 ) , with a latent space living in the representations of the group. We present our architecture and several experimental results on jets at the LHC and find it outperforms graph and convolutional neural network baseline models on several compression, reconstruction, and anomaly detection metrics. We also demonstrate the advantage of such an equivariant model in analyzing the latent space of the autoencoder, which can improve the explainability of potential anomalies discovered by such ML models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Tropical extreme droughts drive long-term increase in atmospheric CO 2 growth rate variability

The terrestrial carbon sink slows the accumulation of carbon dioxide (CO 2 ) in the atmosphere by absorbing roughly 30% of anthropogenic CO 2 emissions, but varies greatly from year to year. The resulting variations in the atmospheric CO 2 growth rate (CGR) have been related to tropical temperature and water availability. The apparent sensitivity of CGR to tropical temperature ( ${{{{{{\rm{\gamma }}}}}}}_{{{{{{\rm{CGR}}}}}}}^{{{{{{\rm{T}}}}}}}$ ) has changed markedly over the past six decades, however, the drivers of the observation to date remains unidentified. Here, we use atmospheric observations, multiple global vegetation models and machine learning products to analyze the cause of the sensitivity change. We found that a threefold increase in ${{{{{{\rm{\gamma }}}}}}}_{{{{{{\rm{CGR}}}}}}}^{{{{{{\rm{T}}}}}}}$ emerged due to the long-term changes in the magnitude of CGR variability (i.e., indicated by one standard deviation of CGR; STD CGR ), which increased 34.7% from 1960-1979 to 1985-2004 and subsequently decreased 14.4% in 1997-2016. We found a close relationship (r 2 = 0.75, p < 0.01) between STD CGR and the tropical vegetated area (23°S – 23°N) affected by extreme droughts, which influenced 6-9% of the tropical vegetated surface. A 1% increase in the tropical area affected by extreme droughts led to about 0.14 Pg C yr -1 increase in STD CGR . The historical changes in STD CGR were dominated by extreme drought-affected areas in tropical Africa and Asia, and semi-arid ecosystems. The outsized influence of extreme droughts over a small fraction of vegetated surface amplified the interannual variability in CGR and explained the observed long-term dynamics of ${{{{{{\rm{\gamma }}}}}}}_{{{{{{\rm{CGR}}}}}}}^{{{{{{\rm{T}}}}}}}$ .

54 ENVIRONMENTAL SCIENCES↗

Revisiting trends in the exchange current for hydrogen evolution

Nørskov and collaborators proposed a simple kinetic model to explain the volcano relation for the hydrogen evolution reaction on transition metal surfaces such that j 0 = k 0 f(ΔG H ) where j 0 is the exchange current density, f(ΔG H ) is a function of the hydrogen adsorption free energy ΔG H as computed from density functional theory, and k 0 is a universal rate constant. Herein, focusing on the hydrogen evolution reaction in acidic medium, we revisit the original experimental data and find that the fidelity of this kinetic model can be significantly improved by invoking metal-dependence on k 0 such that the logarithm of k 0 linearly depends on the absolute value of ΔG H . Here, we further confirm this relationship using additional experimental data points obtained from a critical review of the available literature. Our analyses show that the new model decreases the discrepancy between calculated and experimental exchange current density values by up to four orders of magnitude. Furthermore, we show the model can be further improved using machine learning and statistical inference methods that integrate additional material properties.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

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↗

The quenching of galaxies, bulges, and disks since cosmic noon

Here, we present an analysis of the quenching of star formation in galaxies, bulges, and disks throughout the bulk of cosmic history, from z = 2 – 0. We utilise observations from the Sloan Digital Sky Survey and the Mapping Nearby Galaxies at Apache Point Observatory survey at low redshifts. We complement these data with observations from the Cosmic Assembly Near-Infrared Deep Extragalactic Legacy Survey at high redshifts. Additionally, we compare the observations to detailed predictions from the LGalaxies semi-analytic model. To analyse the data, we developed a machine learning approach utilising a Random Forest classifier. We first demonstrate that this technique is extremely effective at extracting causal insight from highly complex and inter-correlated model data, before applying it to various observational surveys. Our primary observational results are as follows: at all redshifts studied in this work, we find bulge mass to be the most predictive parameter of quenching, out of the photometric parameter set (incorporating bulge mass, disk mass, total stellar mass, and B/T structure). Moreover, we also find bulge mass to be the most predictive parameter of quenching in both bulge and disk structures, treated separately. Hence, intrinsic galaxy quenching must be due to a stable mechanism operating over cosmic time, and the same quenching mechanism must be effective in both bulge and disk regions. Despite the success of bulge mass in predicting quenching, we find that central velocity dispersion is even more predictive (when available in spectroscopic data sets). In comparison to the LGalaxies model, we find that all of these observational results may be consistently explained through quenching via preventative ‘radio-mode’ active galactic nucleus feedback. Furthermore, many alternative quenching mechanisms (including virial shocks, supernova feedback, and morphological stabilisation) are found to be inconsistent with our observational results and those from the literature.

79 ASTRONOMY AND ASTROPHYSICS↗

Photometric redshift-aided classification using ensemble learning

We present SHEEP, a new machine learning approach to the classic problem of astronomical source classification, which combines the outputs from the XGBoost, LightGBM, and CatBoost learning algorithms to create stronger classifiers. A novel step in our pipeline is that prior to performing the classification, SHEEP first estimates photometric redshifts, which are then placed into the data set as an additional feature for classification model training; this results in significant improvements in the subsequent classification performance. SHEEP contains two distinct classification methodologies: (i) Multi-class and (ii) one versus all with correction by a meta-learner. We demonstrate the performance of SHEEP for the classification of stars, galaxies, and quasars using a data set composed of SDSS and WISE photometry of 3.5 million astronomical sources. The resulting F1 -scores are as follows: 0.992 for galaxies; 0.967 for quasars; and 0.985 for stars. In terms of the F1-scores for the three classes, SHEEP is found to outperform a recent RandomForest-based classification approach using an essentially identical data set. Our methodology also facilitates model and data set explainability via feature importances; it also allows the selection of sources whose uncertain classifications may make them interesting sources for follow-up observations.

79 ASTRONOMY AND ASTROPHYSICS↗

Novel principal component analysis tool based on python for analysis of complex spectra of time-of-flight secondary ion mass spectrometry

Time-of-flight secondary ion mass spectrometry (ToF-SIMS) is a powerful surface analysis tool, which can simultaneously provide elemental, isotopic, and molecular information with part per million (ppm) sensitivity. However, each spectrum may be composed of hundreds of ion signals, which makes the spectra data complex. Principal component analysis (PCA) is a multivariate analysis technique that has been widely used to figure out the variances among samples in ToF-SIMS spectra data analysis and is showing great success in the explanation of complex ToF-SIMS spectra. So far, several software tools have been developed for PCA of ToF-SIMS spectra; however, none of them are freely available. Such a situation leads to some difficulties in extending applications of PCA to various research fields. More importantly, it has long been challenging for common researchers to understand PCA plots and extract chemical differences among samples. In this work, we developed a new and flexible software tool (named “advanced spectra pca toolbox”) based on python for PCA of complex ToF-SIMS spectra along with an easy-to-read manual. It can generate data analysis reports automatically to explain chemical differences among samples, allowing less experienced researchers to easily understand tricky PCA results. Moreover, it is expandable and compatible with artificial intelligence/machine learning functions. Pure goethite and different lignin adsorbed goethite samples were used as a model system to demonstrate our new software tool, proving that our software tool can be readily used in complex spectra data processing. Our new software tool is open-source, convenient, flexible, and expandable. We expect this open-source tool will benefit the ToF-SIMS community.

47 OTHER INSTRUMENTATION↗

Generating Mixed Patterns of Residential Segregation: An Evolutionary Approach

The Schelling model of residential segregation has demonstrated that even the slightest preference for neighbors of the same race can be amplified into community-wide segregation. However, these models are unable to simulate mixed, coexisting patterns of segregation and integration, which have been seen to exist in cities. Using evolutionary model discovery we demonstrate how including social factors beyond racial bias when modeling relocation behavior enables the emergence of strongly mixed patterns. Our results indicate that the emergence of mixed patterns is better explained by multiple factors influencing the decision to relocate; the most important being the interaction of nonlinear, rapidly diminishing racial bias with a recent, historical tendency to move. Additionally, preference for less isolated neighborhoods or preference for neighborhoods with longer residing neighbors may produce weaker mixed patterns. Finally, this work highlights the importance of exploring the influence of multiple hypothesized factors of decision making, and their interactions, within agent rules, when studying emergent outcomes generated by agent-based models of complex social systems.

97 MATHEMATICS AND COMPUTING↗

Novel experimental probes of QCD in SIDIS and e + e - annihilation (Final Technical Report)

The research addressed with this award seeks to advance our understanding of the structure and dynamics underlying the properties of visible matter. In our current understanding the nucleons (protons and neutrons) are not fundamental but are comprised of quarks and gluons, which are collectively called partons. A static quark picture fails to explain the properties of the nucleons, such as their mass and intrinsic spin, which are thought to emerge dynamically from the quark-gluon interactions via the strong force. In this work, novel observables employing correlations of particles produced in the scattering of high energy electrons off protons at the CLAS12 experiment at Jefferson Lab were analyzed to probe quark-gluon interactions. The ultimate goal of this line of inquiry is to be able to describe the properties of protons and neutrons from first principles, similar to how studying the hydrogen atom has led to the formulation of the theory of Quantum Electrodynamics. Because quarks cannot be observed directly but only as part of more complex composite particles, a smaller, complimentary part of this work was the analysis of particle production in electron-positron annihilation at the Belle II experiment to understand the production of particles from initial quarks. This takes advantage of the fact that in e + e - annihilation the initial quark dynamics is known, unlike in the scattering of nucleons. Several new applications using Machine Learning algorithms for event tagging and reconstruction to support this program were developed as part of this award. In addition, we had a significant role in the development of the physics program for the future Electron-Ion Collider, which is a new collider to be build in the US within the next decade.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Cloud drop number concentrations over the western North Atlantic Ocean: seasonal cycle, aerosol interrelationships, and other influential factors

Cloud drop number concentrations (N d ) over the western North Atlantic Ocean (WNAO) are generally highest during the winter (DJF) and lowest in summer (JJA), in contrast to aerosol proxy variables (aerosoloptical depth, aerosol index, surface aerosol mass concentrations, surface cloud condensation nuclei (CCN) concentrations) that generally peak inspring (MAM) and JJA with minima in DJF. Using aircraft, satellite remote sensing, ground-based in situ measurement data, and reanalysis data, we characterize factors explaining the divergent seasonal cycles and furthermore probe into factors influencing N d on seasonal timescales. The results can be summarized well by features most pronounced in DJF, including features associated with cold-air outbreak (CAO) conditions such as enhanced values of CAO index, planetary boundary layer height (PBLH),low-level liquid cloud fraction, and cloud-top height, in addition to winds aligned with continental outflow. Data sorted into high- and low-N d days in each season, especially in DJF, revealed that all of these conditions were enhanced on the high-N d days, including reduced sea level pressure and stronger wind speeds. Although aerosols may be more abundant in MAM and JJA, the conditions needed to activate those particles into cloud droplets are weaker than in colder months, which is demonstrated by calculations of the strongest (weakest) aerosol indirect effects in DJF (JJA) based on comparing N d to perturbations in four different aerosol proxy variables (total and sulfate aerosol optical depth, aerosol index, surface mass concentration of sulfate). We used three machine learning models and up to 14 input variables to infer about most influential factors related to N d for DJF and JJA, with the best performance obtained with gradient-boosted regression tree (GBRT) analysis. The model results indicated that cloud fraction was the most important input variable, followed by some combination (depending on season) of CAO index and surface mass concentrations of sulfate and organic carbon. Future work is recommended to further understand aspects uncovered here such as impacts of free tropospheric aerosol entrainment on clouds, degree of boundary layer coupling, wet scavenging, and giant CCN effects on aerosol–N d relationships, updraft velocity, and vertical structure of cloud properties such as adiabaticity that impact the satellite estimation of N d .

54 ENVIRONMENTAL SCIENCES↗

Study of MHD Stability, Active Mode Control, and Disruption Avoidance in NSTX-U

Learning how to achieve sustained operation at high plasma pressure is an important goal of the National Spherical Torus Upgrade (NSTX-U) program and is critically important to any future fusion power plant. This research addresses high priority stability and control issues outlined by the NSTX-U Program. Reduced models of the stabilization physics learned continue to be created and tested against actual plasma operation in NSTX, NSTX-U, and other tokamak devices. This research is forming physics-based models for disruption forecasting that aims to unify results across different fusion machines for more confident extrapolation of the techniques studied to future devices including ITER. The Disruption Event Characterization and Forecasting (DECAF) code was expanded to analyze data from the TCV tokamak, including real-time MHD analysis signals. The presence of a mode and likelihood of its m/n numbers can be tracked, as well as the mode frequency, although a limitation at low frequency does not yet allow tracking modes to the locking point in TCV. Initial analyses of NSTX-U operation were conducted using the DECAF code. The timing of sawtooth oscillations was investigated taking advantage of the more extensive L-mode operation of the device in its first run campaign. An initial investigation of thermal collapses in NSTX-U was also conducted with specific comparison to the Greenwald density limit. A local island power balance model to explain the density limit in tokamaks has been tested with NSTX data and is being evaluated as a potential disruption forecaster in the DECAF code, presently showing similarities to the global Greenwald density evolution and limit. A module was written for the DECAF code that generalizes magnetic spectrogram analysis of MHD modes. This new portable code, written in Python, discriminates the toroidal mode number of instabilities and can track each mode as a DECAF event, monitoring the evolution of the mode through bifurcation and locking points. This is a critical new capability to allow disruption event characterization for general tokamak databases.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Predicting tumor cell line response to drug pairs with deep learning

The National Cancer Institute drug pair screening effort against 60 well-characterized human tumor cell lines (NCI-60) presents an unprecedented resource for modeling combinational drug activity. We present a computational model for predicting cell line response to a subset of drug pairs in the NCI-ALMANAC database. Based on residual neural networks for encoding features as well as predicting tumor growth, our model explains 94% of the response variance. While our best result is achieved with a combination of molecular feature types (gene expression, microRNA and proteome), we show that most of the predictive power comes from drug descriptors. To further demonstrate value in detecting anticancer therapy, we rank the drug pairs for each cell line based on model predicted combination effect and recover 80% of the top pairs with enhanced activity. We present promising results in applying deep learning to predicting combinational drug response. Our feature analysis indicates screening data involving more cell lines are needed for the models to make better use of molecular features.

59 BASIC BIOLOGICAL SCIENCES↗

Impacts of air pollutants from fire and non-fire emissions on the regional air quality in Southeast Asia

Severe haze events in Southeast Asia caused by particulate pollution have become more intense and frequent in recent years, degrading air quality, threatening human health, and interrupting economic and societal activities. Widespread biomass burning activities are a major source of severe haze events in Southeast Asia. On the other hand, particulate pollutants from human activities other than biomass burning also play an important role in degrading air quality in Southeast Asia. In this study, numerical simulations have been conducted using the Weather Research and Forecasting (WRF) model coupled with a chemistry component (WRF-Chem) to quantitatively examine the contributions of aerosols emitted from fire (i.e., biomass burning) versus non-fire (including fossil fuel combustion, road and industrial dust, land use, and land change, etc.) sources to the degradation of air quality and visibility over Southeast Asia. These simulations cover a time period from 2002 to 2008 and were respectively driven by emissions from: (a) fossil fuel burning only, (b) biomass burning only, and (c) both fossil fuel and biomass burning. Across ASEAN 50 cities, these model results reveal that 39&thinsp;% of observed low visibility days can be explained by either fossil fuel burning or biomass burning emissions alone, a further 20&thinsp;% by fossil fuel burning alone, a further 8&thinsp;% by biomass burning alone, and a further 5&thinsp;% by a combination of fossil fuel burning and biomass burning. The remaining 28&thinsp;% of observed low visibility days remain unexplained, likely due to emissions sources that have not been accounted for. Further analysis of 24-hr PM 2.5 Air Quality Index (AQI) indicates that comparing to the simulated result of the case with stand-alone non-fire emissions, the case with coexisting fire and non-fire PM 2.5 can substantially increase the chance of AQI being in the moderate or unhealthy pollution level from 23&thinsp;% to 34&thinsp;%. The premature mortality among major Southeast Asian cities due to degradation of air quality by particulate pollutants is estimated to increase from ~&thinsp;4110 per year in 2002 to ~&thinsp;6540 per year in 2008. In addition, we demonstrate the importance of certain missing non-fire anthropogenic aerosol sources including anthropogenic fugitive and industrial dusts in causing urban air quality degradation. An exploratory experiment of using machine learning algorithms to forecasting the occurrence of haze events in Singapore is also demonstrated in this study. All these results suggest that besides minimizing biomass burning activities, an effective air pollution mitigation policy for Southeast Asia needs to consider controlling emissions from non-fire anthropogenic sources.

Lee, Hsiang-He (ORCID:0000000251407324)↗

Interpreting machine learning prediction of fire emissions and comparison with FireMIP process-based models

Annual burned areas in the United States have increased 2-fold during the past decades. With more large fires resulting in more emissions of fine particulate matter, an accurate prediction of fire emissions is critical for quantifying the impacts of fires on air quality, human health, and climate. This study aims to construct a machine learning (ML) model with game-theory interpretation to predict monthly fire emissions over the contiguous US (CONUS) and to understand the controlling factors of fire emissions. The optimized ML model is used to diagnose the process-based models in the Fire Modeling Intercomparison Project (FireMIP) to inform future development. Results show promising performance for the ML model, Community Land Model (CLM), and Joint UK Land Environment Simulator-Interactive Fire And Emission Algorithm For Natural Environments (JULES-INFERNO) in reproducing the spatial distributions, seasonality, and interannual variability of fire emissions over the CONUS. Regional analysis shows that only the ML model and CLM simulate the realistic interannual variability of fire emissions for most of the subregions (r >0.95 for ML and r =0.14~0.70 for CLM), except for Mediterranean California, where all the models perform poorly (r =0.74 for ML and r <0.30 for the FireMIP models). Regarding seasonality, most models capture the peak emission in July over the western US. However, all models except for the ML model fail to reproduce the bimodal peaks in July and October over Mediterranean California, which may be explained by the smaller wind speeds of the atmospheric forcing data during Santa Ana wind events and limitations in model parameterizations for capturing the effects of Santa Ana winds on fire activity. Furthermore, most models struggle to capture the spring peak in emissions in the southeastern US, probably due to underrepresentation of human effects and the influences of winter dryness on fires in the models. As for extreme events, both the ML model and CLM successfully reproduce the frequency map of extreme emission occurrence but overestimate the number of months with extremely large fire emissions. Comparing the fire PM 2.5 emissions from the ML model with process-based fire models highlights their strengths and uncertainties for regional analysis and prediction and provides useful insights into future directions for model improvements.

54 ENVIRONMENTAL SCIENCES↗

LC-Opt: Benchmarking Reinforcement Learning and Agentic AI for End-to-End Liquid Cooling Optimization in Data Centers

Liquid cooling is critical for thermal management in high-density data centers with the rising AI workloads. However, machine learning-based controllers are essential to unlock greater energy efficiency and reliability, promoting sustainability. We present LC-Opt, a Sustainable Liquid Cooling (LC) benchmark environment, for reinforcement learning (RL) control strategies in energy-efficient liquid cooling of high-performance computing (HPC) systems. Built on the baseline of a high-fidelity digital twin of Oak Ridge National Lab's Frontier Supercomputer cooling system, LC-Opt provides detailed Modelica-based end-to-end models spanning site-level cooling towers to data center cabinets and server blade groups. RL agents optimize critical thermal controls like liquid supply temperature, flow rate, and granular valve actuation at the IT cabinet level, as well as cooling tower (CT) setpoints through a Gymnasium interface, with dynamic changes in workloads. This environment creates a multi-objective real-time optimization challenge balancing local thermal regulation and global energy efficiency, and also supports additional components like a heat recovery unit (HRU). We benchmark centralized and decentralized multi-agent RL approaches, demonstrate policy distillation into decision and regression trees for interpretable control, and explore LLM-based methods that explain control actions in natural language through an agentic mesh architecture designed to foster user trust and simplify system management. LC-Opt democratizes access to detailed, customizable liquid cooling models, enabling the ML community, operators, and vendors to develop sustainable data center liquid cooling control solutions.

Naug, Avisek [Hewlett Packard Enterprise]↗

Exocortex Network for AI-Augmented Human-Led Scientific Expedition

AI advances in science can be viewed along two main directions with a fluid boundary: enhancing efficiency through automation and smart tools to accelerate tasks that humans can already perform; and enabling exploration into uncharted territories and potentially toward AGI. These advances manifest in the AI cognitive core through the development and explainability of foundation models; in the physical embodiment of instruments and facilities; and in the integrated agency of AI workflows exemplified by the science exocortex. To address the role of humans in this evolving landscape, in this Perspective, we suggest a third direction: the development of personalized agents that form human-centered networks, supporting both efficiency and exploration while ensuring that AI remains aligned with human vision.

97 MATHEMATICS AND COMPUTING↗

Scalable Technologies Achieving Risk-Informed Condition-Based Predictive Maintenance Enhancing the Economic Performance of Operating Nuclear Power Plants

The primary objective of the research presented in this report is to develop scalable technologies that are deployable across plant assets and across the nuclear fleet to achieve risk-informed predictive maintenance (PdM) strategies at commercial nuclear power plants (NPPs). Over the years, the nuclear fleet has relied on labor-intensive and time-consuming preventive maintenance (PM) programs, driving up operation and maintenance (O&M) costs to achieve high capacity factors. A well-constructed risk-informed PdM approach for an identified plant asset has been developed in this research, taking advantage of advancements in data analytics, machine learning (ML), artificial intelligence (AI), physics-informed modeling, and visualization. These technologies would allow commercial NPPs to reliably transition from current labor-intensive PM programs to a technology driven PdM program, eliminating unnecessary O&M costs. The work presented in the report is being developed as part of a collaborative research effort between Idaho National Laboratory and Public Service Enterprise Group Nuclear, LLC. This report (1) reflects the results of work by LWRS Program researchers with PSEG, Nuclear LLC-owned Salem and Hope Creek Nuclear Power Plants; (2) presents utilization of circulating water system (CWS) heterogeneous data and fault modes from both the Salem and Hope Creek nuclear power plant sites to develop salient fault signatures associated with each fault mode; (3) describes the integration of component-level predictive models into a robust system-level model enabled by the federated-transfer learning; (4) describes the development of physics-informed model of circulating water pump and motor; (5) develops a scalable risk and economic model; and (6) outlines the development of a user-centric visualization application. The outcomes presented in this report lays the foundation and provides a much-needed technical basis to focus on explainability and trustworthiness of ML and AI-based technologies, as part of future research.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Explainable AI classification for parton density theory

Quantitatively connecting properties of parton distribution functions (PDFs, or parton densities) to the theoretical assumptions made within the QCD analyses which produce them has been a longstanding problem in HEP phenomenology. To confront this challenge, we introduce an ML-based explainability framework, XAI4PDF, to classify PDFs by parton flavor or underlying theoretical model using ResNet-like neural networks (NNs). By leveraging the differentiable nature of ResNet models, this approach deploys guided backpropagation to dissect relevant features of fitted PDFs, identifying x-dependent signatures of PDFs important to the ML model classifications. By applying our framework, we are able to sort PDFs according to the analysis which produced them while constructing quantitative, human-readable maps locating the x regions most affected by the internal theory assumptions going into each analysis. This technique expands the toolkit available to PDF analysis and adjacent particle phenomenology while pointing to promising generalizations.

Artificial Intelligence↗