Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “interpretable 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 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Optimizing Classifiers for Radionuclide Identification

Identifying threat nuclear materials is a critical for the prevention of acts of nuclear terrorism on the homeland. For this purpose, many radionuclide identification devices are deployed in the field. However, these will not necessarily be in the hands of non-experts, therefore these devices need to provide ready-made answers for the personnel in the field. This is where advanced algorithms are employed to both interpret the data and provide the identification of the nuclear material being interrogated. We took a machine learning approach to identification, by using training and validation data sets to create and optimize classifiers which determine which radionuclide is consistent with the data. The classifiers investigated were the Random Forest, Decision Tree, Support Vector Machine, and XG Boost and their performance was judged using the F1 Score for both hyperparameter tuning and comparison. In the end, we found out that the Random Forest Classifier worked the best based off the F1 Score they got which was 0.98.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2025

Spatiotemporal Downscaling Model for Solar Irradiance Forecast Using Nearest-Neighbor Random Forest and Gaussian Process

Accurate solar photovoltaic (PV) capacity estimation requires high-resolution, site-specific solar irradiance data to account for localized variability. However, global datasets, such as the National Solar Radiation Database (NSRDB), provide regional averages that fail to capture the fine-scale fluctuations critical for large-scale grid integration. This limitation is particularly relevant in the context of increasing distributed energy resources (DERs) penetration, such as rooftop PV. Additionally, it is critical to the implementation of the U.S. Federal Energy Regulatory Commission (FERC) Order 2222, which facilitates DER participation in U.S. bulk power markets. To address this challenge, this study evaluates Nearest-Neighbor Random Forest (NNRF) and Nearest-Neighbor Gaussian Process (NNGP) models for spatiotemporal downscaling of global solar irradiance data. By leveraging historical irradiance and meteorological data, these models incorporate spatial, temporal, and feature-based correlations to enhance local irradiance predictions. The NNRF model, a machine-learning approach, prioritizes computational efficiency and predictive accuracy, while the NNGP model offers a level of interpretability and prediction uncertainty by numerically quantifying correlations and dependencies in the data. Model validation was conducted using day-ahead predictions. The results showed that the average Goodness of Fit (GoF) of the NNRF model of 90.61% across all eight sites outperformed the GoF of the NNGP of 85.88%. Additionally, the computational speed of NNRF was 2.5 times faster than the NNGP. Finally, the NNGP displayed polynomial scaling while the NNRF scaled linearly with increasing number of nearest neighbors. Additional validation of the model on five sites in Puerto Rico further confirmed the superiority of the NNRF model over the NNGP model. These findings highlight the robustness and computational efficiency of NNRF for large-scale solar irradiance downscaling, making it a strong candidate for improving PV capacity estimation and real-time electricity market integration for DERs.

Asiedu, Shadrack (ORCID:0009000646004826)↗
DOE OSTI2025

Advancing molecular machine learning representations with stereoelectronics-infused molecular graphs

Molecular representation is a critical element in our understanding of the physical world and the foundation for modern molecular machine learning. Previous molecular machine learning models have used strings, fingerprints, global features and simple molecular graphs that are inherently information-sparse representations. However, as the complexity of prediction tasks increases, the molecular representation needs to encode higher fidelity information. This work introduces a new approach to infusing quantum-chemical-rich information into molecular graphs via stereoelectronic effects, enhancing expressivity and interpretability. Learning to predict the stereoelectronics-infused representation with a tailored double graph neural network workflow enables its application to any downstream molecular machine learning task without expensive quantum-chemical calculations. We show that the explicit addition of stereoelectronic information substantially improves the performance of message-passing two-dimensional machine learning models for molecular property prediction. We show that the learned representations trained on small molecules can accurately extrapolate to much larger molecular structures, yielding chemical insight into orbital interactions for previously intractable systems, such as entire proteins, opening new avenues of molecular design. Finally, we have developed a web application (simg.cheme.cmu.edu) where users can rapidly explore stereoelectronic information for their own molecular systems.

Boiko, Daniil A↗
DOE OSTI2025

Cartesian equivariant representations for learning and understanding molecular orbitals

Qualitative and quantitative orbital properties such as bonding/antibonding character, localization, and orbital energies are critical to how chemists understand reactivity, catalysis, and excited-state behavior. Despite this, representations of orbitals in deep learning models have been very underdeveloped relative to representations of molecular geometries and Hamiltonians. Here, we apply state-of-the-art equivariant deep learning architectures to the task of assigning global labels to orbitals, namely energies characterizations, given the molecular coefficients from Hartree–Fock or density functional theory. The architecture we have developed, the Cartesian Equivariant Orbital Network (CEONET), shows how molecular orbital coefficients are readily featurized as equivariant node features common to all graph-based machine-learned potentials. We find that CEONET performs well at predicting difficult quantitative labels such as the orbital energy and orbital entropy. Furthermore, we find that the CEONET representation provides an intuitive latent space for differentiating orbital character for the qualitative assignment of e.g. bonding or antibonding character. In addition to providing a useful representation for further integrating deep learning with electronic structure theory, we expect CEONET to be useful for automatizing and interpreting the results of advanced electronic structure methods such as complete active space self-consistent field theory. In particular, the ability of CEONET to infer multireference character via the orbital entropy paves the way toward the machine-learned selection of active spaces.

chemical reactions↗
DOE OSTI2018

Energy-based descriptors to rapidly predict hydrogen storage in metal–organic frameworks

The low volumetric density of hydrogen is a critical limitation to its use as a transportation fuel. Filling a fuel tank with nanoporous materials, such as metal–organic frameworks (MOFs), could greatly improve the deliverable capacity of these tanks if appropriate materials could be found. Yet, since MOFs can be made from many combinations of metal nodes, organic linkers, and functional groups, the design space of possible MOFs is enormous. Experimental characterization of thousands of MOFs is infeasible, and even conventional molecular simulations can be prohibitively expensive for large databases. In this work, we have developed a data-driven approach to accelerate materials screening and learn structure–property relationships. We report new descriptors for gas adsorption in MOFs derived from the energetics of MOF–guest interactions. Using the bins of an energy histogram as features, we trained a sparse regression model to predict gas uptake in multiple MOF databases to an accuracy within 3 g L -1 . The interpretable model parameters indicate that a somewhat weak attraction between hydrogen and the framework is ideal for cryogenic storage and release. Our machine learning method is more than three orders of magnitude faster than conventional molecular simulations, enabling rapid exploration of large numbers of MOFs. As a case study, we applied the method to screen a database of more than 50 000 experimental MOF structures. We experimentally validated one of the top candidates identified from the accelerated screening, MFU-4l. This material demonstrated a hydrogen deliverable capacity of 47 g L -1 (54 g L -1 simulated) when operating at storage conditions of 77 K, 100 bar and delivery at 160 K, 5 bar.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Deep learning of dynamically responsive chemical Hamiltonians with semiempirical quantum mechanics

Conventional machine-learning (ML) models in computational chemistry learn to directly predict molecular properties using quantum chemistry only for reference data. While these heuristic ML methods show quantum-level accuracy with speeds several orders of magnitude faster than traditional quantum chemistry methods, they suffer from poor extensibility and transferability; i.e., their accuracy degrades on large or new chemical systems. Incorporating quantum chemistry frameworks into the ML models directly solves this problem. Here we take the structure of semiempirical quantum mechanics (SEQM) methods to construct dynamically responsive Hamiltonians. SEQM methods use empirical parameters fitted to experimental properties to construct reduced-order Hamiltonians, facilitating much faster calculations than ab initio methods but with compromised accuracy. By replacing these static parameters with machine-learned dynamic values inferred from the local environment, we greatly improve the accuracy of the SEQM methods. Trained on molecular energies and atomic forces, these dynamically generated Hamiltonian parameters show a strong correlation with atomic hybridization and bonding. Trained with only about 60,000 small organic molecular conformers, the resulting model retains interpretability, extensibility, and transferability when testing on much larger chemical systems and predicting various molecular properties. Overall, this work demonstrates the virtues of incorporating physics-based descriptions with ML to develop models that are simultaneously accurate, transferable, and interpretable.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2016

Theory-Guided Machine Learning in Materials Science

Materials scientists are increasingly adopting the use of machine learning tools to discover hidden trends in data and make predictions. Applying concepts from data science without foreknowledge of their limitations and the unique qualities of materials data, however, could lead to errant conclusions. The differences that exist between various kinds of experimental and calculated data require careful choices of data processing and machine learning methods. Here, we outline potential pitfalls involved in using machine learning without robust protocols. We address some problems of overfitting to training data using decision trees as an example, rational descriptor selection in the field of perovskites, and preserving physical interpretability in the application of dimensionality reducing techniques such as principal component analysis. We show how proceeding without the guidance of domain knowledge can lead to both quantitatively and qualitatively incorrect predictive models.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Exploration of Electrochemical Reactions at Organic–Inorganic Halide Perovskite Interfaces via Machine Learning in In Situ Time-of-Flight Secondary Ion Mass Spectrometry

The instability of hybrid organic–inorganic perovskite (HOIP) devices is one of the significant challenges preventing commercialization. Exploring these phenomena is severely limited by the complexity of the intrinsic electrochemistry of HOIPs, the presence of multiple volatile and mobile ionic species, and the possible role of environmentally induced reactions at surfaces and triple-phase junctions. Here, in situ studies of the electrochemistry of methylammonium lead bromide perovskite with the Au electrode interface are reported via light- and voltage-dependent time-of-flight secondary ion mass spectrometry (ToF-SIMS) imaging of lateral perovskite heterostructures. While ToF-SIMS allows for the visualization of the chemical composition along the surface and its evolution with light and electrical bias, the interpretation of the multidimensional data obtained is often limited due to strong correlations between chemical signatures and the need to track multiple peaks at once. Here, a machine learning workflow combining the Hough transform and non-negative matrix factorization and no-negative tensor decomposition is developed to avoid this limitation and extract salient features of associated chemical changes and to separate the light- and voltage-dependent dynamics. Finally, combining these in situ characterizations and the machine learning workflow provides comprehensive information on the chemical nature of moving species, ion accumulation, and interfacial electrochemical reactions in HOIP devices.

36 MATERIALS SCIENCE↗
DOE OSTI2024

When ancient numerical demons meet physics-informed machine learning: adjoint-based gradients for implicit differentiable modeling

Recent advances in differentiable modeling, a genre of physics-informed machine learning that trains neural networks (NNs) together with process-based equations, have shown promise in enhancing hydrological models' accuracy, interpretability, and knowledge-discovery potential. Current differentiable models are efficient for NN-based parameter regionalization, but the simple explicit numerical schemes paired with sequential calculations (operator splitting) can incur numerical errors whose impacts on models' representation power and learned parameters are not clear. Implicit schemes, however, cannot rely on automatic differentiation to calculate gradients due to potential issues of gradient vanishing and memory demand. Here we propose a “discretize-then-optimize” adjoint method to enable differentiable implicit numerical schemes for the first time for large-scale hydrological modeling. The adjoint model demonstrates comprehensively improved performance, with Kling–Gupta efficiency coefficients, peak-flow and low-flow metrics, and evapotranspiration that moderately surpass the already-competitive explicit model. Therefore, the previous sequential-calculation approach had a detrimental impact on the model's ability to represent hydrological dynamics. Furthermore, with a structural update that describes capillary rise, the adjoint model can better describe baseflow in arid regions and also produce low flows that outperform even pure machine learning methods such as long short-term memory networks. The adjoint model rectified some parameter distortions but did not alter spatial parameter distributions, demonstrating the robustness of regionalized parameterization. Despite higher computational expenses and modest improvements, the adjoint model's success removes the barrier for complex implicit schemes to enrich differentiable modeling in hydrology.

58 GEOSCIENCES↗
DOE OSTI2022

Composition-transferable machine learning potential for LiCl-KCl molten salts validated by high-energy x-ray diffraction

Unraveling the liquid structure of multicomponent molten salts is challenging due to the difficulty in conducting and interpreting high-temperature diffraction experiments. Here, motivated by this challenge, we developed composition-transferable Gaussian approximation potential (GAP) for molten LiCl-KCl. A DFT-SCAN accurate GAP is active-learned from only ~1100 training configurations drawn from 10 unique mixture compositions enriched with metadynamics. The GAP-computed structures show strong agreement across high-energy x-ray diffraction experiments, including for a eutectic not explicitly included in model training, thereby opening the possibility of composition discovery.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2024

Streamlining latent spaces in machine learning using moment pooling

Many machine learning applications involve learning a latent representation of data, which is often high-dimensional and difficult to directly interpret. In this work, we propose “moment pooling,” a natural extension of deep sets networks which drastically decreases the latent space dimensionality of these networks while maintaining or even improving performance. Moment pooling generalizes the summation in deep sets to arbitrary multivariate moments, which enables the model to achieve a much higher effective latent dimensionality for a fixed learned latent space dimension. We demonstrate moment pooling on the collider physics task of quark/gluon jet classification by extending energy flow networks (EFNs) to moment EFNs. We find that moment EFNs with latent dimensions as small as 1 perform similarly to ordinary EFNs with higher latent dimension. This small latent dimension allows for the internal representation to be directly visualized and interpreted, which in turn enables the learned internal jet representation to be extracted in closed form. Published by the American Physical Society 2024

Gambhir, Rikab (ORCID:0000000251080448)↗
DOE OSTI2026

Generative AI in Supply Chain Management: Applications, Challenges, and Future Directions

Supply chain management (SCM) is undergoing rapid transformation due to increasing global complexity, demand volatility, and operational disruptions. Generative Artificial Intelligence (GenAI) has emerged as a powerful paradigm capable of synthesizing data, simulating operational scenarios, and enabling adaptive decision-making across supply chain networks. This paper presents a survey of GenAI’s role in SCM, focusing on its applications in predictive analytics, autonomous logistics, and fraud detection. Unlike traditional AI systems that rely primarily on predictive analytics, GenAI models, including large language models, generative adversarial networks, and diffusion-based architectures, enable the creation of synthetic supply chain scenarios and autonomous optimization strategies. This survey provides (1) a taxonomy of GenAI techniques for supply chain applications, (2) a comparative analysis of generative AI approaches with traditional machine learning, reinforcement learning, and blockchain-based methods, and (3) a discussion of key challenges such as data privacy, interpretability, and integration with legacy enterprise systems. Furthermore, we outline open research problems and propose directions for future research toward autonomous, resilient, and sustainable AI-driven supply chains.

15 - GEOTHERMAL ENERGY↗
DOE OSTI2021

Physics-Informed Machine Learning from Observations for Clouds, Convection, and Precipitation Parameterizations and Analysis

Fusing learning from Earth observations from space and from the ground (e.g., ARM) with newly developed interpretable and generalizable physics-informed model structures to improve the parameterization of clouds and convection and advance the simulation, understanding and analysis of hydrological extreme events. This covers the prescribed foci of (2) predictive modeling using a hierarchy of models and (3) insight gleaned from complex data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Use of Longitudinal Serum Analysis and Machine Learning to Develop a Classifier for Cancer Early Detection

Early detection of solid tumors through a simple screening process, such as the proteomic analysis of biofluids, has the potential to significantly alter the management and outcomes of cancers. The application of advanced targeted proteomics measurements and data analysis strategies to uniformly collected serum or plasma samples would enable longitudinal studies of cancer risk, progression, and response to therapy that have the potential to significantly reduce cancer burden in general. In this article, we describe a generalizable workflow combining robust, multiplexed targeted proteomics measurements applied to longitudinal samples from the Department of Defense Serum Repository with a Random Forest machine learning method for developing and initially evaluating the performance of candidate biomarker panels for early detection of cancers. The effectiveness of this approach was demonstrated in a cohort of 175 head and neck squamous cell carcinoma patients. The outlined protocols include methods for sample preparation, instrument analysis, and data analysis and interpretation using this workflow.

Longitudinal analysis, machine learning, cancer, e↗
DOE OSTI2021

CrossCheck: Rapid, Reproducible, and Interpretable Model Evaluation

Evaluation beyond aggregate performance metrics, e.g. F1-score, is crucial to both establish an appropriate level of trust in machine learning models and identify future model improvements. In this paper we demonstrate CrossCheck, an interactive visualization tool for rapid crossmodel comparison and reproducible error analysis. We describe the tool and discuss design and implementation details. We then present three use cases (named entity recognition, reading comprehension, and clickbait detection) that show the benefits of using the tool for model evaluation. CrossCheck allows data scientists to make informed decisions to choose between multiple models, identify when the models are correct and for which examples, investigate whether the models are making the same mistakes as humans, evaluate models’ generalizability and highlight models’ limitations, strengths and weaknesses. Furthermore, CrossCheck is implemented as a Jupyter widget, which allows rapid and convenient integration into data scientists’ model development workflows.

Arendt, Dustin L.↗
DOE OSTI2020

Neurotransmitter networks in mouse prefrontal cortex are reconfigured by isoflurane anesthesia

This study quantified eight, small molecule neurotransmitters collected simultaneously from prefrontal cortex of C57BL/6J mouse (n=23) during wakefulness and during isoflurane anesthesia (1.3%). Using isoflurane anesthesia as an independent variable enabled evaluation of the hypothesis that isoflurane anesthesia differentially alters concentrations of multiple neurotransmitters and their interactions. Machine learning was applied to reveal higher order interactions among neurotransmitters. Using a between-subjects design, microdialysis was performed during wakefulness and during anesthesia. Concentrations (nM) of acetylcholine, adenosine, dopamine, GABA, glutamate, histamine, norepinephrine, and serotonin in the dialysis samples are reported (mean ± SD). Relative to wakefulness, acetylcholine concentration was lower during isoflurane anesthesia (1.254 ± 1.118 versus 0.401 ± 0.134, P=0.009), and concentrations of adenosine (29.456 ± 29.756 versus 101.321 ± 38.603, P<0.001), dopamine (0.0578 ± 0.0384 versus 0.113 ± 0.084, P=0.036), and norepinephrine (0.126 ± 0.080 versus 0.219 ± 0.066, P=0.010) were higher during anesthesia. Isoflurane reconfigured neurotransmitter interactions in prefrontal cortex, and the state of isoflurane anesthesia was reliably predicted by prefrontal cortex concentrations of adenosine, norepinephrine, and acetylcholine. A novel finding to emerge from machine learning analyses is that neurotransmitter concentration profiles in mouse prefrontal cortex undergo functional reconfiguration during isoflurane anesthesia. Adenosine, norepinephrine, and acetylcholine showed high feature importance, supporting the interpretation that interactions among these three transmitters may play a key role in modulating levels of cortical and behavioral arousal.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

A Bayesian Approach to Predict Solubility Parameters

Abstract Solubility is a ubiquitous phenomenon in many aspects of material science. While solubility can be determined by considering the cohesive forces in a liquid via the Hansen solubility parameters (HSP), quantitative structure–property relationship models are often used for prediction, notably due to their low computational cost. Here, gpHSP, an interpretable and versatile probabilistic approach to determining HSP, is reported. Our model is based on Gaussian processes, a Bayesian machine learning approach that provides uncertainty bounds to prediction. gpHSP achieves its flexibility by leveraging a variety of input data, such as SMILES strings, COSMOtherm simulations, and quantum chemistry calculations. gpHSP is built on experimentally determined HSP, including a general solvents set aggregated from the literature, and a polymer set experimentally characterized by this group of authors. In all sets, a high degree of agreement is obtained, surpassing well‐established machine learning methods. The general applicability of gpHSP to miscibility of organic semiconductors, drug compounds, and in general solvents is demonstrated, which can be further extended to other domains. gpHSP is a fast and accurate toolbox, which could be applied to molecular design for solution processing technologies.

Sanchez‐Lengeling, Benjamin↗