Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Machine Learning Models”

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

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

At least 1,117 records · Page 62
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Secure Federated Learning Across Heterogeneous Cloud and High-Performance Computing Resources: A Case Study on Federated Fine-Tuning of LLaMA 2

Federated learning enables multiple data owners to collaboratively train robust machine learning models without transferring large or sensitive local datasets by only sharing the parameters of the locally trained models. Here, in this article, we elaborate on the design of our Advanced Privacy-Preserving Federated Learning (APPFL) framework, which streamlines end-to-end secure and reliable federated learning experiments across cloud computing facilities and high-performance computing resources by leveraging Globus Compute, a distributed function as a service platform, and Amazon Web Services. We further demonstrate the use case of APPFL in fine-tuning an LLaMA 2 7B model using several cloud resources and supercomputers.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

A Machine Learning-Based Method to Estimate Transformer Primary-Side Voltages with Limited Customer-Side AMI Measurements

Distribution control applications such as volt/var optimization, network reconfiguration, and distribution automation require accurate knowledge of the distribution system state. The lack of sufficient sensors on the primary side of distribution networks often limits the accuracy of the control decisions by these applications. The deployment of advanced metering infrastructure (AMI) provides utilities an opportunity to translate the AMI data on the secondary onto the primary so that it can be used as pseudo-measurements to augment the limited existing measurements on the primary. This paper develops a machine learning based approach for estimating service transformer primary-side voltages by using limited secondary-side AMI measurement. The machine learning model is developed by using random forest algorithm. The estimated primary-side voltages can be used by utilities as pseudo-measurements for distribution control applications. The detailed secondary model topology, which is an essential input data for many existing algorithms, is not required for the proposed method. The performance of the proposed method is validated by using AMI measurements from the field and an actual distribution feeder model of San Diego Gas & Electric Company.

advanced metering infrastructure↗
DOE OSTI2024

Enhancing Electron Microscopy Image Classification Using Data Augmentation

Manual labeling for machine learning tasks such as image classification is tedious and labor-intensive; as a result, scientific datasets suitable for deep learning applications are scarce and limited. While data augmentation techniques have shown promise for extending image datasets, very little work has been done to understand the impact of combining multiple augmentation methods sequentially or the limits of their effectiveness when combined. Our work addresses this gap by examining how standard and combinatorial data augmentation affects the performance of machine learning models when trained on small datasets for label classification tasks. For our analysis, we generate single, double and quadruple-augmented datasets for a microscopy image classification task using six standard augmentation methods, and compare the resultant improvements observed in binary classification accuracy with three standard image classification models (DenseNet169, MobileNetV2, ResNet101V2). Our experiments show a non-monotonic relationship between the number of simultaneous augmentation methods and classification accuracy, indicating that there is a trade-off between the degree of augmentation and the model performance. These findings suggest that the optimal number of augmentation methods will vary by domain and use case. We also find that the order in which augmentation methods are applied to a limited dataset matters when combining augmentation schemes, with our use case showing performance differences up to 2.6% when the augmentation order is reversed for double-augmented datasets. Our work offers insights to the limits of data augmentation when working on image classification tasks with limited datasets.

Welsman, Jordan A↗
DOE OSTI2021

A Machine Learning-Based Vulnerability Analysis for Cascading Failures of Integrated Power-Gas Systems

This article proposes a cascading failure simulation (CFS) method and a hybrid machine learning method for vulnerability analysis of integrated power-gas systems (IPGSs). The CFS method is designed to study the propagating process of cascading failures between the two systems, generating data for machine learning with initial states randomly sampled. The proposed method considers generator and gas well ramping, transmission line and gas pipeline tripping, island issue handling and load shedding strategies. Then, a hybrid machine learning model with a combined random forest (RF) classification and regression algorithms is proposed to investigate the impact of random initial states on the vulnerability metrics of IPGSs. Extensive case studies are carried out on three test IPGSs to verify the proposed models and algorithms. Simulation results show that the proposed models and algorithms can achieve high accuracy for the vulnerability analysis of IPGSs.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Strangers in a foreign land: ‘Yeastizing’ plant enzymes

Abstract Expressing plant metabolic pathways in microbial platforms is an efficient, cost‐effective solution for producing many desired plant compounds. As eukaryotic organisms, yeasts are often the preferred platform. However, expression of plant enzymes in a yeast frequently leads to failure because the enzymes are poorly adapted to the foreign yeast cellular environment. Here, we first summarize the current engineering approaches for optimizing performance of plant enzymes in yeast. A critical limitation of these approaches is that they are labour‐intensive and must be customized for each individual enzyme, which significantly hinders the establishment of plant pathways in cellular factories. In response to this challenge, we propose the development of a cost‐effective computational pipeline to redesign plant enzymes for better adaptation to the yeast cellular milieu. This proposition is underpinned by compelling evidence that plant and yeast enzymes exhibit distinct sequence features that are generalizable across enzyme families. Consequently, we introduce a data‐driven machine learning framework designed to extract ‘yeastizing’ rules from natural protein sequence variations, which can be broadly applied to all enzymes. Additionally, we discuss the potential to integrate the machine learning model into a full design‐build‐test cycle.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Systematic modification of functionality in disordered elastic networks through free energy surface tailoring

A combined machine learning–physics–based approach is explored for molecular and materials engineering. Specifically, collective variables, akin to those used in enhanced sampled simulations, are constructed using a machine learning model trained on data gathered from a single system. Through the constructed collective variables, it becomes possible to identify critical molecular interactions in the considered system, the modulation of which enables a systematic tailoring of the system’s free energy landscape. To explore the efficacy of the proposed approach, we use it to engineer allosteric regulation and uniaxial strain fluctuations in a complex disordered elastic network. Its successful application in these two cases provides insights regarding how functionality is governed in systems characterized by extensive connectivity and points to its potential for design of complex molecular systems.

42 ENGINEERING↗
DOE OSTI2023

Building a genome-based understanding of bacterial pH preferences

The environmental preferences of many microbes remain undetermined. This is the case for bacterial pH preferences, which can be difficult to predict a priori despite the importance of pH as a factor structuring bacterial communities in many systems. We compiled data on bacterial distributions from five datasets spanning pH gradients in soil and freshwater systems (1470 samples), quantified the pH preferences of bacterial taxa across these datasets, and compiled genomic data from representative bacterial taxa. While taxonomic and phylogenetic information were generally poor predictors of bacterial pH preferences, we identified genes consistently associated with pH preference across environments. We then developed and validated a machine learning model to estimate bacterial pH preferences from genomic information alone, a model that could aid in the selection of microbial inoculants, improve species distribution models, or help design effective cultivation strategies. More generally, we demonstrate the value of combining biogeographic and genomic data to infer and predict the environmental preferences of diverse bacterial taxa.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Design of intrinsically disordered protein variants with diverse structural properties

Intrinsically disordered proteins (IDPs) perform a broad range of functions in biology, suggesting that the ability to design IDPs could help expand the repertoire of proteins with novel functions. Computational design of IDPs with specific conformational properties has, however, been difficult because of their substantial dynamics and structural complexity. We describe a general algorithm for designing IDPs with specific structural properties. We demonstrate the power of the algorithm by generating variants of naturally occurring IDPs that differ in compaction, long-range contacts, and propensity to phase separate. We experimentally tested and validated our designs and analyzed the sequence features that determine conformations. We show how our results are captured by a machine learning model, enabling us to speed up the algorithm. Our work expands the toolbox for computational protein design and will facilitate the design of proteins whose functions exploit the many properties afforded by protein disorder.

Science & Technology - Other Topics↗
DOE OSTI2025

A high-throughput experimentation platform for data-driven discovery in electrochemistry

Automating electrochemical analyses combined with artificial intelligence is poised to accelerate discoveries in renewable energy sciences and technologies. This study presents an automated high-throughput electrochemical characterization (AHTech) platform as a cost-effective and versatile tool for rapidly assessing liquid analytes. The Python-controlled platform combines a liquid handling robot, potentiostat, and customizable microelectrode bundles for diverse, reproducible electrochemical measurements in microtiter plates, minimizing chemical consumption and manual effort. To showcase the capability of AHTech, we screened a library of 180 small molecules as electrolyte additives for aqueous zinc metal batteries, generating data for training machine learning models to predict Coulombic efficiencies. Key molecular features governing additive performance were elucidated using Shapley Additive exPlanations and Spearman’s correlation, pinpointing high-performance candidates like cis-4-hydroxy-d-proline, which achieved an average Coulombic efficiency of 99.52% over 200 cycles. The workflow established herein is highly adaptable, offering a powerful framework for accelerating the exploration and optimization of extensive chemical spaces across diverse energy storage and conversion fields.

Lin, Dian-Zhao [Johns Hopkins University, Baltimor↗
DOE OSTI2025

Predicting fusion ignition at the National Ignition Facility with physics-informed deep learning

Here, an inertial confinement fusion experiment, carried out at the National Ignition Facility, has achieved ignition by generating fusion energy exceeding the laser energy that drove the experiment. Prior to the experiment, a generative machine learning model that combines radiation hydrodynamics simulations, deep learning, experimental data, and Bayesian statistics was used to predict, with a probability greater than 70%, that ignition was the most likely outcome for this shot.

Spears, Brian K. [Lawrence Livermore National Labo↗
DOE OSTI2025

Microbial vitamin biosynthesis links gut microbiota dynamics to chemotherapy toxicity

ABSTRACT Dose-limiting toxicities pose a major barrier to cancer treatment. While preclinical studies show that the gut microbiota influences and is influenced by anticancer drugs, data from patients paired with careful side effect monitoring remains limited. Here, we investigate capecitabine (CAP)-microbiome interactions through longitudinal metagenomic sequencing of stool from 56 advanced colorectal cancer patients. CAP significantly altered the gut microbiome, enriching for menaquinol (vitamin K2) biosynthesis genes. Transposon library screens, targeted gene deletions, and media supplementation revealed that menaquinol biosynthesis protectsEscherichia colifrom drug toxicity. Stool menaquinol gene and metabolite levels were associated with decreased peripheral sensory neuropathy. Machine learning models trained in this cohort predicted toxicities in an independent cohort. Taken together, these results suggest treatment-associated increases in microbial vitamin biosynthesis serve a chemoprotective role for bacterial and host cells. Further, our findings provide a foundation for in-depth mechanistic dissection, human intervention studies, and extension to other cancer treatments. IMPORTANCE Side effects are common during the treatment of cancer. The trillions of microbes found within the human gut are sensitive to anticancer drugs, but the effects of treatment-induced shifts in gut microbes for side effects remain poorly understood. We profiled gut microbes in colorectal cancer patients treated with capecitabine and carefully monitored side effects. We observed a marked expansion in genes for producing vitamin K2 (menaquinone). Vitamin K2 rescued gut bacterial growth and was associated with decreased side effects in patients. We then used information about gut microbes to develop a predictive model of drug toxicity that was validated in an independent cohort. These results suggest that treatment-associated increases in bacterial vitamin production protect both bacteria and host cells from drug toxicity, providing new opportunities for intervention and motivating the need to better understand how dietary intake and bacterial production of micronutrients like vitamin K2 influence cancer treatment outcomes.

Microbiology↗
DOE OSTI2022

Geochemistry and Multiomics Data Differentiate Streams in Pennsylvania Based on Unconventional Oil and Gas Activity

Unconventional oil and gas (UOG) extraction is increasing exponentially around the world, as new technological advances have provided cost-effective methods to extract hard-to-reach hydrocarbons. While UOG has increased the energy output of some countries, past research indicates potential impacts in nearby stream ecosystems as measured by geochemical and microbial markers. Here, we utilized a robust data set that combines 16S rRNA gene amplicon sequencing (DNA), metatranscriptomics (RNA), geochemistry, and trace element analyses to establish the impact of UOG activity in 21 sites in northern Pennsylvania. These data were also used to design predictive machine learning models to determine the UOG impact on streams. We identified multiple biomarkers of UOG activity and contributors of antimicrobial resistance within the order Burkholderiales. Furthermore, we identified expressed antimicrobial resistance genes, land coverage, geochemistry, and specific microbes as strong predictors of UOG status. Of the predictive models constructed (n = 30), 15 had accuracies higher than expected by chance and area under the curve values above 0.70. The supervised random forest models with the highest accuracy were constructed with 16S rRNA gene profiles, metatranscriptomics active microbial composition, metatranscriptomics active antimicrobial resistance genes, land coverage, and geochemistry (n = 23). The models identified the most important features within those data sets for classifying UOG status. These findings identified specific shifts in gene presence and expression, as well as geochemical measures, that can be used to build robust models to identify impacts of UOG development.

16S rRNA↗
DOE OSTI2024

The expressivity of classical and quantum neural networks on entanglement entropy

Abstract Analytically continuing the von Neumann entropy from Rényi entropies is a challenging task in quantum field theory. While then-th Rényi entropy can be computed using the replica method in the path integral representation of quantum field theory, the analytic continuation can only be achieved for some simple systems on a case-by-case basis. In this work, we propose a general framework to tackle this problem using classical and quantum neural networks with supervised learning. We begin by studying several examples with known von Neumann entropy, where the input data is generated by representing$${\text {Tr}}\rho _A^n$$ Tr ρ A n with a generating function. We adopt KerasTuner to determine the optimal network architecture and hyperparameters with limited data. In addition, we frame a similar problem in terms of quantum machine learning models, where the expressivity of the quantum models for the entanglement entropy as a partial Fourier series is established. Our proposed methods can accurately predict the von Neumann and Rényi entropies numerically, highlighting the potential of deep learning techniques for solving problems in quantum information theory.

Physics↗
DOE OSTI2026

Reconstruction of atmospheric neutrinos in DUNE’s horizontal-drift far-detector module

This paper reports on the capabilities in reconstructing and identifying atmospheric neutrino interactions in one of the Deep Underground Neutrino Experiment’s (DUNE) far detector modules, a liquid argon time projection chamber (LArTPC) with horizontal drift (FD-HD) of ionization electrons. The reconstruction is based upon the workflow developed for DUNE’s long-baseline oscillation analysis, with some necessary machine-learning models’ retraining and the addition of features relevant only to atmospheric neutrinos such as the neutrino direction reconstruction. Where relevant, the impact of the detection of the charged particles of the hadronic system is emphasized, and comparisons are carried out between the case when lepton-only information is considered in the reconstruction (as is the case for many neutrino oscillation experiments), versus when all particles identified in the LArTPC were included. Three neutrino direction reconstruction methods have been developed and studied for the atmospheric analyses: using lepton-only information, using all reconstructed particles, and using only correlations from reconstructed hits. The results indicate that incorporating more than just lepton information significantly improves the resolution of both neutrino direction and energy reconstruction. The angle reconstruction algorithms developed in this work result in no strong dependence on particle direction for reconstruction efficiencies or neutrino flavor identification. This comprehensive review of the reconstruction of atmospheric neutrinos in DUNE’s FD-HD LArTPC is the first step towards developing a first neutrino oscillation sensitivity analysis, which will ready DUNE for its first measurements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

How Do Visual Explanations Foster End Users' Appropriate Trust in Machine Learning?

We investigated the effects of different visual explanations on users' trust in machine learning classification. We proposed three forms of visual explanations of a classification based on identifying relevant training instances. We conducted a user study to evaluate these visual explanations as well as a no explanation condition. We measured users' trust of a classifier, quantified the effects of these three forms of explanations, and assessed the changes in users' trust. We found that participants trust a classifier appropriately when an explanation is available. The combination of human, classification algorithm and understandable explanation makes better decisions than the classifier and human alone.This work advances the state-of-the-art closer to building trust-able machine learning models and informs the design and appropriate use of automated systems.

Yang, Fumeng↗
DOE OSTI2020

Characterization and identification of HPC applications at leadership computing facility

High Performance Computing (HPC) is an important method for scientific discovery via large-scale simulation, data analysis, or artificial intelligence. Leadership-class supercomputers are expensive, but essential to run large HPC applications. The Petascale era of supercomputers began in 2008, with the first machines achieving performance in excess of one petaflops, and with the advent of new supercomputers in 2021 (e.g., Aurora, Frontier), the Exascale era will soon begin. However, the high theoretical computing capability (i.e., peak FLOPS) of a machine is not the only meaningful target when designing a supercomputer, as the resources demand of applications varies. A deep understanding of the characterization of applications that run on a leadership supercomputer is one of the most important ways for planning its design, development and operation. In order to improve our understanding of HPC applications, user demands and resource usage characteristics, we perform correlative analysis of various logs for different subsystems of a leadership supercomputer. This analysis reveals surprising, sometimes counter-intuitive patterns, which, in some cases, conflicts with existing assumptions, and have important implications for future system designs as well as supercomputer operations. For example, our analysis shows that while the applications spend significant time on MPI, most applications spend very little time on file I/O. Combined analysis of hardware event logs and task failure logs show that the probability of a hardware FATAL event causing task failure is low. Combined analysis of control system logs and file I/O logs reveals that pure POSIX I/O is used more widely than higher level parallel I/O. Based on holistic insights of the application gained through combined and co-analysis of multiple logs from different perspectives and general intuition, we engineer features to "fingerprint" HPC applications. We use t-SNE (a machine learning technique for dimensionality reduction) to validate the explainability of our features and finally train machine learning models to identify HPC applications or group those with similar characteristic. To the best of our knowledge, this is the first work that combines logs on file I/O, computing, and inter-node communication for insightful analysis of HPC applications in production.

Liu, Zhengchun↗
DOE OSTI2020

Toward Large-Scale Image Segmentation on Summit

Semantic segmentation of images is an important computer vision task that emerges in a variety of application domains such as medical imaging, robotic vision and autonomous vehicles to name a few. While these domain-specific image analysis tasks involve relatively small image sizes (~ 10 2 × 10 2 ), there are many applications that need to train machine learning models on image data with extents that are orders of magnitude larger (~10 4 × 10 4 ). Training deep neural network (DNN) models on large extent images is extremely memory-intensive and often exceeds the memory limitations of a single graphical processing unit, a hardware accelerator of choice for computer vision workloads. Here, an efficient, sample parallel approach to train U-Net models on large extent image data sets is presented. Its advantages and limitations are analyzed and near-linear strong-scaling speedup demonstrated on 256 nodes (1536 GPUs) of the Summit supercomputer. Using a single node of the Summit supercomputer, an early evaluation of a recently released model parallel framework called GPipe is demonstrated to deliver ~ 2X speedup in executing a U-Net model with an order of magnitude larger number of trainable parameters than reported before. Performance bottlenecks for pipelined training of U-Net models are identified and mitigation strategies to improve the speedups are discussed. Together, these results open up the possibility of combining both approaches into a unified scalable pipelined and data parallel algorithm to efficiently train U-Net models with very large receptive fields on data sets of ultra-large extent images.

Seal, Sudip↗