SEARCH · Engineering Papers

Results for “explainability”

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.

SourceResearch connectionFrom yearThrough year

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

At least 73 records · Page 4

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Synoptic Meteorology Explains Temperate Forest Carbon Uptake

Abstract While substantial attention has been paid to the effects of both global climate oscillations and local meteorological conditions on the interannual variability of ecosystem carbon exchange, the relationship between the interannual variability of synoptic meteorology and ecosystem carbon exchange has not been well studied. Here we use a clustering algorithm to identify a summertime cyclonic precipitation system northwest of the Great Lakes to determine (a) the association at a daily scale between the occurrence of this system and the local meteorology and net ecosystem exchange at three Great Lakes region forested eddy covariance sites and (b) the association between the seasonal prevalence of this system and the summertime net ecosystem exchange of these sites. We find that temperature, in addition to precipitation and cloud cover, is an important explanatory factor for the suppression of net ecosystem productivity that occurs during these cyclonic events in this region. In addition, the prevalence of this cyclonic system can explain a significant proportion of the interannual variability in summertime forest ecosystem exchange in this region. This explanatory power is not due to a simple accumulation of low‐productivity days that cooccur with this meteorological event, but rather a broader association between the frequency of these events and several aspects of prevailing seasonal conditions. This work demonstrates the usefulness of conceptualizing meteorology in terms of synoptic systems for explaining the interannual variability of regional carbon fluxes.

Randazzo, Nina A.↗

DOE OSTI2022

Using an Explainable Machine Learning Approach to Characterize Earth System Model Errors: Application of SHAP Analysis to Modeling Lightning Flash Occurrence

Abstract Computational models of the Earth System are critical tools for modern scientific inquiry. Efforts toward evaluating and improving errors in representations of physical and chemical processes in these large computational systems are commonly stymied by highly nonlinear and complex error behavior. Recent work has shown that these errors can be effectively predicted using modern Artificial Intelligence (A.I.) techniques. In this work, we go beyond these previous studies to apply an explainable A.I. technique to not only predict model errors but also move toward understanding the underlying reasons for successful error prediction. We use XGBoost classification trees and SHapley Additive exPlanations analysis to explore the errors in the prediction of lightning occurrence in the NASA Goddard Earth Observing System model, a widely used Earth System Model. This explainable error prediction system can effectively predict the model error and indicates that the errors are strongly related to convective processes and the characteristics of the land surface.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Detection of Forced Change Within Combined Climate Fields Using Explainable Neural Networks

Abstract Assessing forced climate change requires the extraction of the forced signal from the background of climate noise. Traditionally, tools for extracting forced climate change signals have focused on one atmospheric variable at a time, however, using multiple variables can reduce noise and allow for easier detection of the forced response. Following previous work, we train artificial neural networks to predict the year of single‐ and multi‐variable maps from forced climate model simulations. To perform this task, the neural networks learn patterns that allow them to discriminate between maps from different years—that is, the neural networks learn the patterns of the forced signal amidst the shroud of internal variability and climate model disagreement. When presented with combined input fields (multiple seasons, variables, or both), the neural networks are able to detect the signal of forced change earlier than when given single fields alone by utilizing complex, nonlinear relationships between multiple variables and seasons. We use layer‐wise relevance propagation, a neural network explainability tool, to identify the multivariate patterns learned by the neural networks that serve as reliable indicators of the forced response. These “indicator patterns” vary in time and between climate models, providing a template for investigating inter‐model differences in the time evolution of the forced response. This work demonstrates how neural networks and their explainability tools can be harnessed to identify patterns of the forced signal within combined fields.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2015

Climate variation explains a third of global crop yield variability

Many studies have examined the role of mean climate change in agriculture, but an understanding of the influence of inter-annual climate variations on crop yields in different regions remains elusive. We use detailed crop statistics time series for B13,500 political units to examine how recent climate variability led to variations in maize, rice, wheat and soybean crop yields worldwide. While some areas show no significant influence of climate variability, in substantial areas of the global breadbaskets, 460% of the yield variability can be explained by climate variability. Globally, climate variability accounts for roughly a third (B32–39%) of the observed yield variability. Our study uniquely illustrates spatial patterns in the relationship between climate variability and crop yield variability, highlighting where variations in temperature, precipitation or their interaction explain yield variability. We discuss key drivers for the observed variations to target further research and policy interventions geared towards buffering future crop production from climate variability

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2019

Reliable and explainable machine-learning methods for accelerated material discovery

Abstract Despite ML’s impressive performance in commercial applications, several unique challenges exist when applying ML in materials science applications. In such a context, the contributions of this work are twofold. First, we identify common pitfalls of existing ML techniques when learning from underrepresented/imbalanced material data. Specifically, we show that with imbalanced data, standard methods for assessing quality of ML models break down and lead to misleading conclusions. Furthermore, we find that the model’s own confidence score cannot be trusted and model introspection methods (using simpler models) do not help as they result in loss of predictive performance (reliability-explainability trade-off). Second, to overcome these challenges, we propose a general-purpose explainable and reliable machine-learning framework. Specifically, we propose a generic pipeline that employs an ensemble of simpler models to reliably predict material properties. We also propose a transfer learning technique and show that the performance loss due to models’ simplicity can be overcome by exploiting correlations among different material properties. A new evaluation metric and a trust score to better quantify the confidence in the predictions are also proposed. To improve the interpretability, we add a rationale generator component to our framework which provides both model-level and decision-level explanations. Finally, we demonstrate the versatility of our technique on two applications: (1) predicting properties of crystalline compounds and (2) identifying potentially stable solar cell materials. We also point to some outstanding issues yet to be resolved for a successful application of ML in material science.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Training calibration-based counterfactual explainers for deep learning models in medical image analysis

The rapid adoption of artificial intelligence methods in healthcare is coupled with the critical need for techniques to rigorously introspect models and thereby ensure that they behave reliably. This has led to the design of explainable AI techniques that uncover the relationships between discernible data signatures and model predictions. In this context, counterfactual explanations that synthesize small, interpretable changes to a given query while producing desired changes in model predictions have become popular. This under-constrained, inverse problem is vulnerable to introducing irrelevant feature manipulations, particularly when the model’s predictions are not well-calibrated. Hence, in this paper, we propose the TraCE (training calibration-based explainers) technique, which utilizes a novel uncertainty-based interval calibration strategy for reliably synthesizing counterfactuals. Given the wide-spread adoption of machine-learned solutions in radiology, our study focuses on deep models used for identifying anomalies in chest X-ray images. Using rigorous empirical studies, we demonstrate the superiority of TraCE explanations over several state-of-the-art baseline approaches, in terms of several widely adopted evaluation metrics. Our findings show that TraCE can be used to obtain a holistic understanding of deep models by enabling progressive exploration of decision boundaries, to detect shortcuts, and to infer relationships between patient attributes and disease severity.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2026

Explainable machine learning for incipient anomaly detection in compact molten salt heat exchanger with overlapping feature distributions

High-temperature molten salt-cooled reactors (MSCRs) are a promising next-generation nuclear technology option, offering efficient power conversion and inherent safety features. However, the reliability of these systems depends on the robust operation of heat exchangers (HXs), which are susceptible to failure due to temperature gradients and channel plugging caused by fluid freezing. Conventional monitoring methods, relying on inlet and outlet measurements, lack the spatial resolution needed to detect early-stage faults. We propose a novel design of a compact salt-to-salt matrix-type HX design consisting of interleaved arrays of parallel tubes, with integrated synthetic fiber optic distributed temperature sensing (DTS) to enable localized detection of incipient faults. To evaluate performance of this design, we generate high-fidelity synthetic data using heat transfer computational modeling to simulate channel plugging, and introduce sensor noise for realistic modeling of measurements. The dataset comprises of 97% normal operation and 3% anomaly cases, with each anomaly class representing 1% of the data. These early anomalies result in overlapping temperature profiles between normal and faulty channels, producing a non-separable dataset that challenges traditional classification techniques. We benchmark eight supervised machine learning (ML) models and demonstrate that XGBoost achieves the highest performance. To improve transparency, we develop an explainability framework combining Shapley values and partially ordered sets (POSETs) to quantify and structurally analyze feature importance. This approach identifies both dominant predictors and ambiguous feature relationships, enhancing trust and interpretability. Our results highlight the potential of combining DTS and explainable ML with intelligent feature selection to improve predictive maintenance and ensure operational resilience in advanced nuclear systems.

Prantikos, Konstantinos [Argonne National Laborato↗

DOE OSTI2026

Explainable tokamak-agnostic forecasting of fusion plasma instability via megahertz turbulent fluctuations

Scientific applications of artificial intelligence (AI) often remain limited by device-specific training and unexplained “black-box” approaches, creating fundamental barriers to cross-system generalization. This challenge is critical for nuclear fusion, where future reactors will have limited operational data for AI training. Here, we demonstrate that our neural network, trained solely on megahertz-scale turbulence measurements from one machine (DIII-D), forecasts Type-I edge localized mode (ELM) onsets in a different tokamak (KSTAR) through zero-shot weight transfer following physics-consistent preprocessing without device-specific retraining. Through an explainable AI framework combining gradient-weighted class activation mapping with physics validation, we reveal that our network can internalize physics relationships governing the ELM instabilities rather than memorizing device-specific patterns. The network perceives spatiotemporal features that correlate consistently with independently calculated instability growth rates, magnetohydrodynamic stability limits, and pedestal structure dynamics. Statistical analyses of dimensionally-reduced saliency features reveal the identical triangular features between the saliency representations, instability growth rates, and prediction probability across tokamaks, providing evidence that our forecasting system can show tokamak-agnostic generalization. This work contributes to a foundation for explainable scientific AI systems, where cross-system developments are essential for transcending traditional domain-specific constraints.

AI↗

DOE OSTI2015

Time delay and noise explaining the behaviour of the cell growth in fermentation process

This paper proposes to investigate the interplay between time delay and external noise in explaining the behaviour of the microbial growth in batch fermentation process. Time delay and noise are modelled jointly via stochastic delay differential equations (SDDEs). The typical behaviour of cell concentration in batch fermentation process under this model is investigated. Milstein scheme is applied for solving this model numerically. Simulation results illustrate the effects of time delay and external noise in explaining the lag and stationary phases, respectively for the cell growth of fermentation process.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2015

The Dielectric Breakdown Model applied to explain various morphologies of deposited metallic structures in thin gap metal electro-deposition

The phenomenon of metal electro-deposition in thin-gap geometry leads to very interesting and diverse two dimensional morphologies. This varies from dense ramified growth to thin dendritic projections. In this paper, we have proposed a stochastic model that incorporates such diversity. We carried out thin-gap electro-deposition of Copper and Zinc with varying electrolytic concentrations. A well known model, that until this work was used to explain dielectric breakdown patterns, was employed to explain the variation in deposition morphology with concentration. The sole parameter in the model was varied and the numerically obtained patterns was seen to correlate well with those obtained from electro-deposition. A linear relationship between the parameter and molar concentration was established. The established relationship was then analysed and interpreted.

36 MATERIALS SCIENCE↗

DOE OSTI2026

Explaining System-Level Prognostics with Established Machine Learning Methods

System-level prognostics is crucial for ensuring reliability and enabling predictive maintenance in complex systems with interconnected components. This study presents a framework that integrates data-driven methods to predict the remaining useful life (RUL) of a subsystem under multiple and concurrent faults within a nuclear power plant system with explainable artificial intelligence (XAI). A nuclear power plant (NPP) operation was simulated to model the degradation behavior of NPP components, and four machine learning models—Gradient Boosting Regressor (GBR), Support Vector Regressor (SVR), Fully Connected Neural Network (FCNN), and Long Short-Term Memory (LSTM)—were evaluated for prognostics with a novel system RUL parameter. The LSTM model demonstrated potential superior repeatability, while SHAP (SHapley Additive exPlanations) for explainability provided consistent and trustworthy global explanations. In contrast, LIME (Local Interpretable Model-agnostic Explanations) offered localized interpretability but showed reduced stability for sequential data. Key findings include the interplay between component-level degradation and system-wide performance, with LSTM effectively capturing these dynamics through sequence-level predictions. The XAI techniques enhanced transparency by identifying critical features influencing model predictions and aligning with domain knowledge. Furthermore, this framework has significant implications for improving trust and understanding in predictive maintenance, particularly in safety-critical industries like nuclear energy.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2015

HORIZON-SCALE LEPTON ACCELERATION IN JETS: EXPLAINING THE COMPACT RADIO EMISSION IN M87

It has now become clear that the radio jet in the giant elliptical galaxy M87 must turn on very close to the black hole. This implies the efficient acceleration of leptons within the jet at scales much smaller than feasible by the typical dissipative events usually invoked to explain jet synchrotron emission. Here we show that the stagnation surface, the separatrix between material that falls back into the black hole and material that is accelerated outward forming the jet, is a natural site of pair formation and particle acceleration. This occurs via an inverse Compton pair catastrophe driven by unscreened electric fields within the charge-starved region about the stagnation surface and substantially amplified by a post-gap cascade. For typical estimates of the jet properties in M87, we find excellent quantitive agreement between the predicted relativistic lepton densities and those required by recent high-frequency radio observations of M87. This mechanism fails to adequately fill a putative jet from Sagittarius A{sup *} with relativistic leptons, which may explain the lack of an obvious radio jet in the Galactic center. Finally, this process implies a relationship between the kinetic jet power and the gamma-ray luminosity of blazars, produced during the post-gap cascade.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2017

The variance of the locally measured Hubble parameter explained with different estimators

We study the expected variance of measurements of the Hubble constant, H {sub 0}, as calculated in either linear perturbation theory or using non-linear velocity power spectra derived from N -body simulations. We compare the variance with that obtained by carrying out mock observations in the N-body simulations, and show that the estimator typically used for the local Hubble constant in studies based on perturbation theory is different from the one used in studies based on N-body simulations. The latter gives larger weight to distant sources, which explains why studies based on N-body simulations tend to obtain a smaller variance than that found from studies based on the power spectrum. Although both approaches result in a variance too small to explain the discrepancy between the value of H {sub 0} from CMB measurements and the value measured in the local universe, these considerations are important in light of the percent determination of the Hubble constant in the local universe.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2026

NuGraph2 with explainability: post-hoc explanations for geometric neural network predictions

With the growing popularity of artificial intelligence (AI) used for scientific applications, the ability of attribute a result to a reasoning process from the network is in high demand for robust scientific generalizations to hold. In this work we aim to motivate the need for and demonstrate the use of post-hoc explainability methods when applied to AI methods used in scientific applications. To this end, we introduce explainability add-ons to the existing graph neural network (GNN) for neutrino tagging, NuGraph2. The explanations take the form of a suite of techniques examining the output of the network (node classifications) and the edge connections between them, and probing of the latent space using novel general-purpose tools applied to this network. We show how none of these methods are singularly sufficient to show network ‘understanding’, but together can give insights into the processes used in classification. While these methods are tested on the NuGraph2 application, they can be applied to a broad range of networks, not limited to GNNs. The code for this work is publicly available on GitHub at https://github.com/voetberg/XNuGraph.

Voetberg, Margaret [Fermilab] (ORCID:0009000527154↗

DOE OSTI2025

SetBERT: the deep learning platform for contextualized embeddings and explainable predictions from high-throughput sequencing

MOTIVATION: High-throughput sequencing (HTS) is a modern sequencing technology used to profile microbiomes by sequencing thousands of short genomic fragments from the microorganisms within a given sample. This technology presents a unique opportunity for artificial intelligence to comprehend the underlying functional relationships of microbial communities. However, due to the unstructured nature of HTS data, nearly all computational models are limited to processing DNA sequences individually. This limitation causes them to miss out on key interactions between microorganisms, significantly hindering our understanding of how these interactions influence the microbial communities as a whole. Furthermore, most computational methods rely on post-processing of samples which could inadvertently introduce unintentional protocol-specific bias. RESULTS: Addressing these concerns, we present SetBERT, a robust pre-training methodology for creating generalized deep learning models for processing HTS data to produce contextualized embeddings and be fine-tuned for downstream tasks with explainable predictions. By leveraging sequence interactions, we show that SetBERT significantly outperforms other models in taxonomic classification with genus-level classification accuracy of 95%. Furthermore, we demonstrate that SetBERT is able to accurately explain its predictions autonomously by confirming the biological-relevance of taxa identified by the model. AVAILABILITY AND IMPLEMENTATION: All source code is available at https://github.com/DLii-Research/setbert. SetBERT may be used through the q2-deepdna QIIME 2 plugin whose source code is available at https://github.com/DLii-Research/q2-deepdna.

Ludwig, David W↗

DOE OSTI2024

Genetic and Functional Diversity Help Explain Pathogenic, Weakly Pathogenic, and Commensal Lifestyles in the Genus Xanthomonas

The genus Xanthomonas has been primarily studied for pathogenic interactions with plants. However, besides host and tissue-specific pathogenic strains, this genus also comprises nonpathogenic strains isolated from a broad range of hosts, sometimes in association with pathogenic strains, and other environments, including rainwater. Based on their incapacity or limited capacity to cause symptoms on the host of isolation, nonpathogenic xanthomonads can be further characterized as commensal and weakly pathogenic. This study aimed to understand the diversity and evolution of nonpathogenic xanthomonads compared to their pathogenic counterparts based on their cooccurrence and phylogenetic relationship and to identify genomic traits that form the basis of a life history framework that groups xanthomonads by ecological strategies. We sequenced genomes of 83 strains spanning the genus phylogeny and identified eight novel species, indicating unexplored diversity. While some nonpathogenic species have experienced a recent loss of a type III secretion system, specifically the hrp2 cluster, we observed an apparent lack of association of the hrp2 cluster with lifestyles of diverse species. We performed association analysis on a large data set of 337 Xanthomonas strains to explain how xanthomonads may have established association with the plants across the continuum of lifestyles from commensals to weak pathogens to pathogens. Presence of distinct transcriptional regulators, distinct nutrient utilization and assimilation genes, transcriptional regulators, and chemotaxis genes may explain lifestyle-specific adaptations of xanthomonads.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2017

Self-Interacting Dark Matter Can Explain Diverse Galactic Rotation Curves

The rotation curves of spiral galaxies exhibit a diversity that has been difficult to understand in the cold dark matter (CDM) paradigm. We show that the self-interacting dark matter (SIDM) model provides excellent fits to the rotation curves of a sample of galaxies with asymptotic velocities in the 25-300 km /s range that exemplify the full range of diversity. We assume only the halo concentration-mass relation predicted by the CDM model and a fixed value of the self-interaction cross section. In dark-matter-dominated galaxies, thermalization due to self-interactions creates large cores and reduces dark matter densities. In contrast, thermalization leads to denser and smaller cores in more luminous galaxies and naturally explains the flatness of rotation curves of the highly luminous galaxies at small radii. In conclusion, our results demonstrate that the impact of the baryons on the SIDM halo profile and the scatter from the assembly history of halos as encoded in the concentration-mass relation can explain the diverse rotation curves of spiral galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2023

Making Corgis Important for Honeycomb Classification: Adversarial Attacks on Concept-based Explainability Tools

Methods for model explainability have become increasingly critical for testing the fairness and soundness of deep learning. Concept-based interpretability techniques, which use a small set of human-interpretable concept exemplars in order to measure the influence of a concept on a model's internal representation of input, are an important thread in this line of research. In this work we show that these explainability methods can suffer the same vulnerability to adversarial attacks as the models they are meant to analyze. We demonstrate this phenomenon on two well-known concept-based interpretability methods: TCAV and faceted feature visualization. We show that by leveraging the geometry of the problem and carefully perturbing the examples of the concept that is being investigated, we can radically change the output of the interpretability method. The attacks that we propose can either induce positive interpretations (polka dots are an important concept for a model when classifying zebras) or negative interpretations (stripes are not an important factor in identifying images of a zebra). Our work highlights the fact that in safety-critical applications, there is need for security around not only the machine learning pipeline but also the model interpretation process.

Brown, Davis R.↗