Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Explanability”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Joint explanation of the B → π K puzzle and the B → K ν ν ¯ excess

In light of the recent branching fraction measurement of the B + → K + ν ν ¯ decay by Belle II and its poor agreement with the Standard Model expectation, we analyze the effects of an axionlike particle (ALP) in B meson decays. We assume a long-lived ALP with a mass of the order of the pion mass that decays to two photons. We focus on a scenario where the ALP decay length is of the order of meters such that the ALP has a non-negligible probability to decay outside the detector volume of Belle II, mimicking the B + → K + ν ν ¯ signal. Remarkably, such an arrangement is also relevant for the long-standing B → π K puzzle by noting that the measured B 0 → π 0 K 0 and B + → π 0 K + decays could have a B 0 → a K 0 and B + → a K + component, respectively. We also argue based on our results that the required ALP-photon effective coupling belongs to a region of parameter space that can be extensively probed in future beam dump experiments like SHiP. Published by the American Physical Society 2025

Altmannshofer, W. (ORCID:0000000316212561)↗
DOE OSTI2025

Multiscale Explanation of the Missing Gallium Vacancy in Gallium Arsenide

Irradiation of gallium arsenide (GaAs) produces immobile vacancies and mobile interstitials. Yet, after decades of experimental investigation, the immobile Ga vacancy continues to evade detection, raising the question: where is the Ga vacancy? Static first-principles calculations predict a Ga vacancy should be readily observed. We find that short-time dynamical evolution of primary defects is the key to explaining this conundrum. Using a dynamical multiscale atomistically informed device engineering (AIDE) method, we discover that during the initial displacement damage, the Ga vacancy (3-/2-) defect level pins the Fermi level near the midgap, producing oppositely charged vacancies and interstitials. Driven by Coulomb attraction, fast As interstitials preferentially annihilate Ga vacancies. The Ga vacancy population plummets below detectable limits—and the now unpinned Fermi level recovers—before being experimentally observed. This dynamical model solves the mystery of the missing Ga vacancy and reveals the importance of a multiscale approach to explore the dynamical chemical behavior in experimentally inaccessible short-time regimes.

Diaz, Leopoldo [Sandia National Laboratories (SNL-↗
DOE OSTI2021

Plateau Phenomenon in Gradient Descent Training of RELU Networks: Explanation, Quantification, and Avoidance

The ability of neural networks to provide ‘best in class’ approximation across a wide range of applications is well-documented. Nevertheless, the powerful expressivity of neural networks comes to naught if one is unable to effectively train (choose) the parameters defining the network. In general, neural networks are trained by gradient descent type optimization methods,a stochastic variant thereof. In practice, such methods result in the loss function decreases rapidly at the beginning of training but then, after a relatively small number of steps, significantly slow down. The loss may even appear to stagnate over the period of a large number of epochs, only to then suddenly start to decrease fast again for no apparent reason. This so-called plateau phenomenon manifests itself in many learning tasks. The present work aims to identify and quantify the root causes of plateau phenomenon.analysis is carried out in the setting of univariate ReLU networks. No assumptions are made on the number of neurons relative to the number of training data, and our results hold for both the lazy and adaptive regimes. Here, the main findings are: plateaux correspond to periods during which activation patterns remain constant, where activation pattern refers to the number of data points that activate a given neuron; quantification of convergence of the gradient flow dynamics; and, characterization stationary points in terms solutions of local least squares regression lines over subsets of the training data. Based on these conclusions, we propose a new iterative training method, the Active Neuron Least Squares (ANLS), characterised by the explicit adjustment of the activation pattern at each step, which is designed to enable a quick exit from a plateau. Illustrative numerical examples are included throughout.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Biota Modeling in EPA’s Preliminary Remediation Goal and Dose Compliance Concentration Calculators for Use in EPA Superfund Risk Assessment: Explanation of Intake Rate Derivation, Transfer Factor Compilation, and Mass Loading Factor Sources

The Preliminary Remediation Goal (PRG) and Dose Compliance Concentration (DCC) calculators are screening level risk assessment tools that set forth the Environmental Protection Agency’s (EPA) recommended approaches and currently available risk assessment guidance for response actions at Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) sites, commonly known as Superfund. The environmental screening levels derived by the PRG and DCC calculators are used to identify isotopes contributing the highest risk and dose as well as establish preliminary remediation goals. Each calculator has residential gardening and subsistence farmer exposure scenarios that model transfer of contaminants from soil and water into various types of biota (crops and animal products). New publications of human intake rates of biota; farm animal intakes of water, soil, and fodder; and soil to plant interactions require updates be implemented into the PRG and DCC calculators. Recent improvements in the biota modeling for these calculators include newly derived biota intake rates, enhanced soil mass loading factors (MLFs), and more comprehensive soil to plant transfer factors (BV’s) and soil to tissue transfer factors (TFs) for animals. New biota have been added in both the produce and animal products categories that greatly improve the accuracy and utility of the PRG and DCC calculators and encompass greater geographic diversity on a national and international scale.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Boson Fermion Nucleus Explanation for Violation of Parity in the Radioactive Decay of Cobalt-60: Monograph #12

In 1957, Chien-Shuing Wu subjected cobalt-60 to supercooled temperatures and a strong magnetic field while measuring radioactive decay. This experiment showed violation of parity— a long-held, fundamental precept in physics stating that nuclear radioactive-decay emission flux will not vary in the solid angle about the nucleus. During Wu’s experiment, beta particles (electrons) exhibited asymmetry, preferentially exiting the cobalt-60 nuclei in the polar region opposite the applied magnetic field. Gamma-ray emissions were anisotropic, preferring to exit the nucleus around the equatorial region. The boson fermion nucleus (BFN) explains the origins of these observed phenomena, where the nuclear structure is pinned in the strong magnetic field with minimal motion in supercooled conditions, while beta particles and gamma rays are emitted from specific locations within the nuclear structure.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Boson Fermion Nucleus Explanation for Fission Fragment Anisotropy: Monograph #13

Since 1952, anisotropy has been observed in the paths of fission fragments following the assimilation of an incident neutron or photon in the nucleus. Fission fragment flux favors certain angles relative to the path of the incident particle, constituting anisotropy. The amount of anisotropy is dependent on the energy of the incident particle. Fission fragment anisotropy can be explained by the disruption of nuclear structure combined with the requirement of a specific rotational orientation in theta and phi of that structure relative to the incident neutron or photon flux. A specific and limited area in the nucleus can assimilate the incident particle and induce upset within the structure, followed by scission and fission of the nucleus. The orientation and nature of the structure upset dictates fission fragment trajectory and is responsible for fission fragment anisotropy. This concept can be explained and supported by the theory of boson fermion nucleus (BFN) structure.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Explanation and Application of the Evolving Contact Traction Fields in Shallow Foundation Systems

The present paper provides a qualitative discussion of the evolution of contact traction fields beneath rigid shallow foundations resting on granular materials. A phenomenological similarity is recognized in the measured contact traction fields of rigid footings and at the bases of sandpiles. This observation leads to the hypothesis that the stress distributions are brought about by the same physical phenomena, namely the development of arching effects through force chains and mobilized intergranular friction. A set of semi-empirical equations are suggested for the normal and tangential components of this contact traction based on past experimental measurements and phenomenological assumptions of frictional behaviors at the foundation system scale. These equations are then applied to the prescribed boundary conditions for the analysis of the settlement, resistance, and stress fields in supporting granular materials beneath the footing. A parametric sensitivity study is performed on the proposed modelling method, highlighting solutions to the boundary-value problems in an isotropic, homogeneous elastic half-space.

42 ENGINEERING↗
DOE OSTI2021

SAGE Intrusion Detection System: Sensitivity Analysis Guided Explainability for Machine Learning.

This report details the results of a three-fold investigation of sensitivity analysis (SA) for machine learning (ML) explainability (MLE): (1) the mathematical assessment of the fidelity of an explanation with respect to a learned ML model, (2) quantifying the trustworthiness of a prediction, and (3) the impact of MLE on the efficiency of end-users through multiple users studies. We focused on the cybersecurity domain as the data is inherently non-intuitive. As ML is being using in an increasing number of domains, including domains where being wrong can elicit high consequences, MLE has been proposed as a means of generating trust in a learned ML models by end users. However, little analysis has been performed to determine if the explanations accurately represent the target model and they themselves should be trusted beyond subjective inspection. Current state-of-the-art MLE techniques only provide a list of important features based on heuristic measures and/or make certain assumptions about the data and the model which are not representative of the real-world data and models. Further, most are designed without considering the usefulness by an end-user in a broader context. To address these issues, we present a notion of explanation fidelity based on Shapley values from cooperative game theory. We find that all of the investigated MLE explainability methods produce explanations that are incongruent with the ML model that is being explained. This is because they make critical assumptions about feature independence and linear feature interactions for computational reasons. We also find that in deployed, explanations are rarely used due to a variety of reason including that there are several other tools which are trusted more than the explanations and there is little incentive to use the explanations. In the cases when the explanations are used, we found that there is the danger that explanations persuade the end users to wrongly accept false positives and false negatives. However, ML model developers and maintainers find the explanations more useful to help ensure that the ML model does not have obvious biases. In light of these findings, we suggest a number of future directions including developing MLE methods that directly model non-linear model interactions and including design principles that take into account the usefulness of explanations to the end user. We also augment explanations with a set of trustworthiness measures that measure geometric aspects of the data to determine if the model output should be trusted.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Leveraging large language models to automate the identification of healthcare access barriers for veterans

Objective: To develop and evaluate an automated system for identifying healthcare barriers focusing on transportation issues in veterans’ clinical notes using large language models (LLMs) and to assess the impact of different prompting strategies on classification performance and explanation consistency. Methods: We developed a hybrid system combining pattern matching for templated notes with LLM analysis for free-text notes. Using 2000 manually annotated clinical notes, we compared four prompting strategies (dual-role short, dual-role long, analysis-first, analysis-only) across Mistral-7B and Llama-3.1 models. We evaluated classification performance using standard metrics and assessed explanation consistency through embedding similarity analysis. Results: The analysis-first strategy achieved superior performance, with Mistral-7B reaching an F1 score of 0.914, outperforming traditional machine learning approaches (GBM: 0.786, BERT: 0.811). LLMs demonstrated higher explanation consistency within models (mean cosine similarity 0.887–0.908) compared to cross-model similarities (0.767–0.872). Pattern matching successfully handled 6.7% of templated notes deterministically. Mistral-7B showed greater internal consistency but higher abstention rates compared to Llama-3.1. Conclusion: Requiring LLMs to analyze evidence before classification improves both accuracy and explanation consistency for identifying transportation barriers in clinical notes. This approach enables automated barrier detection at scale while providing clinically relevant explanations, supporting both population-level healthcare planning and individual patient care decisions.

Healthcare access barriers↗
DOE OSTI2025

ACES-GNN: can graph neural network learn to explain activity cliffs?

Graph Neural Networks (GNNs) have revolutionized molecular property prediction by leveraging graph-based representations, yet their opaque decision-making processes hinder broader adoption in drug discovery. This study introduces the Activity-Cliff-Explanation-Supervised GNN (ACES-GNN) framework, designed to simultaneously improve predictive accuracy and interpretability by integrating explanation supervision for activity cliffs (ACs) into GNN training. ACs, defined by structurally similar molecules with significant potency differences, pose challenges for traditional models due to their reliance on shared structural features. By aligning model attributions with chemist-friendly interpretations, the ACES-GNN framework bridges the gap between prediction and explanation. Validated across 30 pharmacological targets, ACES-GNN consistently enhances both predictive accuracy and attribution quality for ACs compared to unsupervised GNNs. Our results demonstrate a positive correlation between improved predictions and accurate explanations, offering a robust and adaptable framework to better understand and interpret ACs. This work underscores the potential of explanation-guided learning to advance interpretable artificial intelligence in molecular modeling and drug discovery.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Explaining machine-learning models for gamma-ray detection and identification

As more complex predictive models are used for gamma-ray spectral analysis, methods are needed to probe and understand their predictions and behavior. Recent work has begun to bring the latest techniques from the field of Explainable Artificial Intelligence (XAI) into the applications of gamma-ray spectroscopy, including the introduction of gradient-based methods like saliency mapping and Gradient-weighted Class Activation Mapping (Grad-CAM), and black box methods like Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP). In addition, new sources of synthetic radiological data are becoming available, and these new data sets present opportunities to train models using more data than ever before. In this work, we use a neural network model trained on synthetic NaI(Tl) urban search data to compare some of these explanation methods and identify modifications that need to be applied to adapt the methods to gamma-ray spectral data. We find that the black box methods LIME and SHAP are especially accurate in their results, and recommend SHAP since it requires little hyperparameter tuning. We also propose and demonstrate a technique for generating counterfactual explanations using orthogonal projections of LIME and SHAP explanations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗