Search NASA⌕ Search

SEARCH · Search NASA

Results for “Generalizable model”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Data-based filtered dissipation rate modelling for multi-modal turbulent combustion: evaluating a priori model generalizability

Manifold-based models offer a computationally efficient alternative to directly transporting the thermochemical state in computational simulations of turbulent reacting flows, projecting the high-dimensional thermochemical state-space onto a low-dimensional manifold. Recent efforts have yielded a manifold-based model applicable to multi-modal combustion, enabling reconstruction of the thermochemical state from solutions to two-dimensional manifold equations in mixture fraction and generalized progress variable that are parameterised by three scalar dissipation rates. In coarse-grained simulations such as Large Eddy Simulation (LES), closure of the multi-modal manifold equations and subfilter variances/covariance requires closure of three filtered scalar dissipation rates. Here, the present work adopts a data-based approach, providing closure for the three filtered scalar dissipation rates via deep neural networks (DNNs). High-fidelity datasets corresponding to an autoigniting n-dodecane jet flame and a bluff body swirl-stabilized confined lifted spray flame of two aviation fuels (Jet-A and C1) with different ignition propensities are leveraged to generate training data that spans a diverse range of thermodynamic conditions and combustion modes, including low- and high-temperature ignition regimes in addition to premixed and nonpremixed behaviour. A final DNN model is trained to enforce inherent physical constraints by learning nonlinear functional transformations of the three filtered scalar dissipation rates. The generalizability of this constrained DNN model is demonstrated a priori via conditional statistics evaluated on the lifted spray flame with C1–a configuration that had not been included in the training data. Excellent DNN agreement with conditional DNS statistics is observed, and integrated gradients are computed to identify the most sensitive input variables. The similarity of the marginal PDFs of the most informative input variables and outputs across configurations are quantified via the Wasserstein metric, demonstrating that data-based models may successfully generalize to unseen parametric conditions so long as the most informative input variables share similar distributions across training and testing datasets.

Data-based modelling↗
DOE OSTI2025

Inferring Plant Acclimation and Improving Model Generalizability With Differentiable Physics‐Informed Machine Learning of Photosynthesis

Net photosynthesis (A N ) is a key component of the global carbon cycle influencing climate feedback over decadal scales. Although plant acclimation to environmental changes can modify A N , traditional vegetation models in Earth system models (ESMs) often rely on plant functional type (PFT)-specific parameterizations or simplified acclimation assumptions limiting generalizability across time, space, and PFTs. In this study, we developed a differentiable photosynthesis model to learn the environmental dependencies of V c,max25 (maximum carboxylation rate at 25°C, representing photosynthetic capacity), as this genre of hybrid physics-informed machine learning can seamlessly train neural networks and process-based equations together. Compared to PFT-specific parameterization of V c,max25 , learning the environment dependencies of key photosynthetic parameters improved model spatiotemporal generalizability. Applying environmental acclimation to V c,max25 led to substantial variations in global mean A N indicating the need to address acclimation in ESMs. The model effectively captured multivariate observations (V c,max25 , A N , and stomatal conductance (g s )) simultaneously with multivariate constraints, improving generalization across space and PFTs. It also learned sensible acclimation relationships of V c,max25 to different environmental conditions. The model explained more than 54%, 57%, and 62% of the variance of A N , g s , and V c,max25 , respectively, presenting a first global-scale spatial test benchmark of A N and g s . These results highlight the potential for differentiable modeling to enhance process-based modules in ESMs and effectively leverage information from large, multivariate data sets.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Language models for the prediction of SARS-CoV-2 inhibitors

The COVID-19 pandemic highlights the need for computational tools to automate and accelerate drug design for novel protein targets. We leverage deep learning language models to generate and score drug candidates based on predicted protein binding affinity. We pre-trained a deep learning language model (BERT) on ∼9.6 billion molecules and achieved peak performance of 603 petaflops in mixed precision. Our work reduces pre-training time from days to hours, compared to previous efforts with this architecture, while also increasing the dataset size by nearly an order of magnitude. For scoring, we fine-tuned the language model using an assembled set of thousands of protein targets with binding affinity data and searched for inhibitors of specific protein targets, SARS-CoV-2 Mpro and PLpro. We utilized a genetic algorithm approach for finding optimal candidates using the generation and scoring capabilities of the language model. Our generalizable models accelerate the identification of inhibitors for emerging therapeutic targets.

Blanchard, Andrew 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 OSTI2024

Review of In Situ Sensing for Directed Energy Deposition for Industrial Part Quality Assessment

As the use additive manufacturing (AM) processes continues to grow in critical industries, improved quality assurance methods are becoming increasingly sought after for qualification and certification of AM components. Traditional nondestructive evaluation of printed components is often unable to supply the required confidence in print quality to justify qualification and certification, but the layer-by-layer nature of AM provides unprecedented opportunities for in situ quality inspection. This document summarizes recent developments in process monitoring research specifically related to Directed Energy Deposition (DED). Particular attention is given to three aspects of the highlighted manuscripts: (1) the type of sensors used, (2) features extracted from each sensor modality, and (3) analysis of extracted features for AM quality assessment. Based on the review of the state-of-the-art, several observations have been made. First, none of the reviewed works have applied their trained models to real part geometries, with many of the works relying on single track experiments, thin-walled structures, and cubes. Similarly, there have not been any works demonstrating model generalizability, i.e., a model trained on data from one build allows for fruitful analysis of data from another build. Many works used machine learning techniques to distinguish different process regimes (i.e., normal, keyholing, lack-of-fusion), but very few papers have investigated stochastic variation in an already “optimized” process. Sensor fusion approaches are also limited in the DED sensing literature, but the few works that have employed such techniques have demonstrated the benefits. Finally, registration of in situ data to the build coordinate system is of paramount importance to producing industrially relevant in situ monitoring systems. Data registration allows direct correlations between process anomalies detected in the process monitoring data to localized departures in part quality, but such techniques are generally lacking in the current literature.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Toward a Generalizable Prediction Model of Molten Salt Mixture Density with Chemistry‐Informed Transfer Learning

Optimally designing applications of molten salts requires knowledge of their thermophysical properties over a wide range of temperatures and compositions. There exist significant gaps in existing databases and this data can be challenging to experimentally measure due to high temperatures, salt corrosivity, and salt hygroscopicity. Existing databases have been used to create Redlich–Kister (RK) models for mixture density showing improved accuracy with respect to ideal mixing assumptions, but these models require subcomponent data measurements for each new system, therefore lacking generality. In order to address generalizability and data sparsity, a transfer learning procedure is proposed to train deep neural networks (DNNs) using a combination of semi‐empirical relationships (RK), data from the thermophysical arm of the molten salt thermal properties database and universal ab initio properties of component mixtures taken from the joint automated repository for various integrated simulations (JARVIS) classical force‐field inspired descriptors database to predict density in molten salts. Herein, it is shown that DNNs predict molten salt density with an r 2 over 0.99 and a mean absolute percentage error under 1%, outperforming alternative methods.

inorganic materials↗
DOE OSTI2021

A cross-study analysis of drug response prediction in cancer cell lines

Abstract To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and drug features. However, most algorithm development efforts have relied on cross-validation within a single study to assess model accuracy. While an essential first step, cross-validation within a biological data set typically provides an overly optimistic estimate of the prediction performance on independent test sets. To provide a more rigorous assessment of model generalizability between different studies, we use machine learning to analyze five publicly available cell line-based data sets: National Cancer Institute 60, ancer Therapeutics Response Portal (CTRP), Genomics of Drug Sensitivity in Cancer, Cancer Cell Line Encyclopedia and Genentech Cell Line Screening Initiative (gCSI). Based on observed experimental variability across studies, we explore estimates of prediction upper bounds. We report performance results of a variety of machine learning models, with a multitasking deep neural network achieving the best cross-study generalizability. By multiple measures, models trained on CTRP yield the most accurate predictions on the remaining testing data, and gCSI is the most predictable among the cell line data sets included in this study. With these experiments and further simulations on partial data, two lessons emerge: (1) differences in viability assays can limit model generalizability across studies and (2) drug diversity, more than tumor diversity, is crucial for raising model generalizability in preclinical screening.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

INSURE: An Information Theory iNspired diSentanglement and pURification modEl for Domain Generalization

Domain Generalization (DG) aims to learn a generalizable model on the unseen target domain by only training on the multiple observed source domains. Although a variety of DG methods have focused on extracting domain-invariant features, the domain-specific class-relevant features have attracted attention and been argued to benefit generalization to the unseen target domain. To take into account the class-relevant domain-specific information, in this paper we propose an Information theory iNspired diSentanglement and pURification modEl (INSURE) to explicitly disentangle the latent features to obtain sufficient and compact (necessary) class-relevant feature for generalization to the unseen domain. Specifically, we first propose an information theory inspired loss function to ensure the disentangled class-relevant features contain sufficient class label information and the other disentangled auxiliary feature has sufficient domain information. Additionally, we further propose a paired purification loss function to let the auxiliary feature discard all the class-relevant information and thus the class-relevant feature will contain sufficient and compact (necessary) class-relevant information. Moreover, instead of using multiple encoders, we propose to use a learnable binary mask as our disentangler to make the disentanglement more efficient and make the disentangled features complementary to each other. We conduct extensive experiments on five widely used DG benchmark datasets including PACS, VLCS, OfficeHome, TerraIncognita, and DomainNet. The proposed INSURE achieves state-of-the-art performance. We also empirically show that domain-specific class-relevant features are beneficial for domain generalization. The code is available at https://github.com/yuxi120407/INSURE .

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics

Large language models have revolutionized artificial intelligence by enabling large, generalizable models trained through self-supervision. This paradigm has inspired the development of scientific foundation models (FMs). However, applying this capability to experimental particle physics is challenging due to the sparse, spatially distributed nature of detector data, which differs dramatically from natural language. This work addresses if an FM for particle physics can scale and generalize across diverse tasks. We introduce a new dataset with more than 11 million particle collision events and a suite of downstream tasks and labeled data for evaluation. We propose a novel self-supervised training method for detector data and demonstrate its neural scalability with models that feature up to 188 million parameters. With frozen weights and task-specific adapters, this FM consistently outperforms baseline models across all downstream tasks. The performance also exhibits robust data-efficient adaptation. Further analysis reveals that the representations extracted by the FM are task-agnostic but can be specialized via a single linear mapping for different downstream tasks.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

A Framework for Assessing Economic and Environmental Trade-offs of Internalized Emission Costs in ERCOT Grid Planning

The power grid is on the cusp of a massive transition driven by three major areas: 1) the growth in demand for electricity, 2) efforts to decarbonize the United States economy, and 3) a desire to mitigate social disparities from the impact of electricity generation on local populations. However, most studies of the electricity sector do not include equity impacts in their models. This study seeks to do so by developing a comprehensive and generalizable model tailored to the Electric Reliability Council of Texas (ERCOT) grid, designed to incorporate the equity impacts of electricity generation in a decarbonized and resilient framework. To integrate equity into our research, we incorporate environmental externalities into our capacity expansion model of ERCOT. Specifically, we factor in intermediate-level local marginal damages of precursor pollutants (NH3, NOx, primary PM2.5, SO2, and VOC) and global pollutant CO2 into the cost of generating electricity. We do this by taking into account county population, county ambient pollution concentration, and generator emission rates. Leveraging open-source modeling tools, such as PowerGenome, pyGRETA, and GenX we construct a county-level model to account for these costs. We integrate these marginal damages into the variable operations and maintenance costs of generators, for both existing and potential future builds. This study’s findings suggest that the value of a dynamic social cost of carbon (SSC) will cover criteria pollutant marginal damages within the ERCOT grid and solar and wind is expected to increase out to 2035. Key metrics evaluated within this research include fuel mix distribution across technologies, transmission and grid infrastructure costs, CO2 emissions, local pollutants marginal damages, and the variation in generation capacity built by the model. These results and framework can be used to support grid decisions that explicitly include distributional and procedural equity within a decarbonized and sustainable grid framework.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Building thermal dynamics modeling with deep transfer learning using a large residential smart thermostat dataset

Understanding thermal dynamics and obtaining the computational model of residential buildings enable its scaled application in energy retrofits, control optimization and decarbonization. In this paper, we present a deep learning approach to model building thermal dynamics with smart thermostat data collected from residential buildings, with the goal to investigate model generalizability. In the first stage, we developed and compared different Deep Learning architectures including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) models and CNN-LSTM to predict indoor air temperature in a multi-step time horizon. In the second stage, we implemented a Transfer Learning (TL) process, which aims to improve the prediction performance on a new set of buildings (targets), exploiting the knowledge of related or similar buildings (sources). Different TL strategies and source model identification methods were investigated. The study showed that the CNN-LSTM performed the best among the architectures compared, with an average Mean Absolute Error (MAE) of 0.26 °C for one-hour-ahead (twelve 5-min future steps) predictions. Furthermore, the results showed that freezing the LSTM layer and fine-tuning the other layers of the CNN-LSTM achieved the best performance among four TL strategies, which further improved the performance with respect to a machine learning approach by 10%, and proving the effectiveness and generalizability of the proposed approach. A comparison of three different source model identification methods showed that randomly selecting source models constrained by similar building characteristics can provide good TL performance while retaining simplicity comparing with other quantitative source identification methods.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Challenges in predicting protein-protein interactions of understudied viruses: Arenavirus-human interactions

Understanding protein-protein interactions (PPIs) between viruses and host organisms is crucial for uncovering infection mechanisms and identifying potential therapeutic targets. The ability to generalize PPI predictive models across understudied viruses presents a significant challenge. In this work, we use arenavirus-human PPIs to illustrate the difficulties associated with model generalization, which are compounded by a lack of both positive and negative data. We employ a Transfer Learning approach to investigate arenavirus-human PPIs by utilizing models trained on better-studied virus-human and human-human PPIs. Additionally, we curate and assess four types of negative sampling datasets to evaluate their impact on model performance. Despite the overall high accuracies (93–99 %) and AUPRC scores (0.8–0.9) appearing promising, further analysis indicates that these performance metrics can be misleading due to data leakage, data bias, and overfitting, especially concerning under-represented viral proteins. We reveal these gaps and assess the impact of data imbalance using standard k-fold cross-validation and Independent Blind Testing with a Balanced Dataset, resulting in a drop in accuracy below 50 %. We propose a viral protein-specific evaluation framework that categorizes viral proteins into majority and minority classes based on their representation in the dataset, enabling comparison of model performance across these groups using balanced accuracies. This framework offers a more robust evaluation of model generalizability, addressing biases inherent in standard evaluation techniques and paving the way for more reliable PPI prediction models for understudied viruses.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

On the Generalizability of Time-of-Flight Convolutional Neural Networks for Noninvasive Acoustic Measurements

Bulk wave acoustic time-of-flight (ToF) measurements in pipes and closed containers can be hindered by guided waves with similar arrival times propagating in the container wall, especially when a low excitation frequency is used to mitigate sound attenuation from the material. Convolutional neural networks (CNNs) have emerged as a new paradigm for obtaining accurate ToF in non-destructive evaluation (NDE) and have been demonstrated for such complicated conditions. However, the generalizability of ToF-CNNs has not been investigated. In this work, we analyze the generalizability of the ToF-CNN for broader applications, given limited training data. We first investigate the CNN performance with respect to training dataset size and different training data and test data parameters (container dimensions and material properties). Furthermore, we perform a series of tests to understand the distribution of data parameters that need to be incorporated in training for enhanced model generalizability. This is investigated by training the model on a set of small- and large-container datasets regardless of the test data. We observe that the quantity of data partitioned for training must be of a good representation of the entire sets and sufficient to span through the input space. The result of the network also shows that the learning model with the training data on small containers delivers a sufficiently stable result on different feature interactions compared to the learning model with the training data on large containers. To check the robustness of the model, we tested the trained model to predict the ToF of different sound speed mediums, which shows excellent accuracy. Furthermore, to mimic real experimental scenarios, data are augmented by adding noise. We envision that the proposed approach will extend the applications of CNNs for ToF prediction in a broader range.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

Sampling Subjective Polygons for Patch-Based Deep Learning Land-Use Classification in Satellite Images

Model generalization remains a key challenge in the analysis of large amounts of heterogeneous satellite image data. One major limiting factor in developing generalizable models, in the context of supervised learning, is the lack of high quality training datasets. A model's capacity to perform well on new data is often inhibited by imbalance and bias in the data that was used for training. This is especially a problem when using convolutional neural networks to classify urban land-use in satellite images. Notable dataset imbalance issues in this application include land-use type imbalance and image scene imbalance. To begin understanding these dataset imbalance problems in more detail, we develop and test a number of sampling methods for generating training image datasets from subjective training polygons for urban land-use classification. We investigate sampling at different point densities as a means to reduce content repetition and therefore content imbalance and bias in the training image dataset.

Arndt, Jacob↗
DOE OSTI2024

Towards Diverse and Representative Global Pretraining Datasets for Remote Sensing Foundation Models

The design of a pretraining dataset is emerging as a critical component for the generality of foundation models. In the remote sensing realm, large volumes of imagery and benchmark datasets exist that can be leveraged to pretrain foundation models, however using this imagery in absence of a well-crafted sampling strategy is inefficient and has the potential to create biased and less generalizable models. Here, we provide a discussion and vision for the curation and assessment of pretraining datasets for remote sensing geospatial foundation models. We highlight the importance of geographic, temporal, and image acquisition diversity and review possible strategies to enable such diversity at global scale. In addition to these characteristics, support for various spatial-temporal pretext tasks within the dataset is also critical. Ultimately, our primary objective is to place emphasis on and draw attention to the data curation stage of the foundation model development pipeline. By doing so, we think it is possible to reduce biases of geospatial foundation models, as well as enable broader generalization to downstream remote sensing tasks and applications.

Arndt, Jacob↗
DOE OSTI2023

Differentiable physics-enabled closure modeling for Burgers’ turbulence

Abstract Data-driven turbulence modeling is experiencing a surge in interest following algorithmic and hardware developments in the data sciences. We discuss an approach using the differentiable physics paradigm that combines known physics with machine learning to develop closure models for Burgers’ turbulence. We consider the one-dimensional Burgers system as a prototypical test problem for modeling the unresolved terms in advection-dominated turbulence problems. We train a series of models that incorporate varying degrees of physical assumptions on an a posteriori loss function to test the efficacy of models across a range of system parameters, including viscosity, time, and grid resolution. We find that constraining models with inductive biases in the form of partial differential equations that contain known physics or existing closure approaches produces highly data-efficient, accurate, and generalizable models, outperforming state-of-the-art baselines. Addition of structure in the form of physics information also brings a level of interpretability to the models, potentially offering a stepping stone to the future of closure modeling.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

SIDDA: SInkhorn Dynamic Domain Adaptation for image classification with equivariant neural networks

Modern neural networks (NNs) often do not generalize well in the presence of a ‘covariate shift’; that is, in situations where the training and test data distributions differ, but the conditional distribution of classification labels given the data remains unchanged. In such cases, NN generalization can be reduced to a problem of learning more robust, domain-invariant features. Domain adaptation (DA) methods include a broad range of techniques aimed at achieving this; however, these methods have struggled with the need for extensive hyperparameter tuning, which then incurs significant computational costs. In this work, we introduce SInkhorn Dynamic Domain Adaptation (SIDDA), an out-of-the-box DA training algorithm built upon the Sinkhorn divergence, that can achieve effective domain alignment with minimal hyperparameter tuning and computational overhead. We demonstrate the efficacy of our method on multiple simulated and real datasets of varying complexity, including simple shapes, handwritten digits, real astronomical observations, and remote sensing data. These datasets exhibit covariate shifts due to noise, blurring, differences between telescopes, and variations in imaging wavelengths. SIDDA is compatible with a variety of NN architectures, and it works particularly well in improving classification accuracy and model calibration when paired with symmetry-aware equivariant NNs (ENNs). We find that SIDDA consistently enhances the generalization capabilities of NNs, achieving up to a ${\approx}40\%$ improvement in classification accuracy on unlabeled target data, while also providing a more modest performance gain of $\lesssim 1\%$ on labeled source data. We also study the efficacy of DA on ENNs with respect to the varying group orders of the dihedral group DN, and find that the model performance improves as the degree of equivariance increases. Finally, if SIDDA achieves proper domain alignment, it also enhances model calibration on both source and target data, with the most significant gains in the unlabeled target domain—achieving over an order of magnitude improvement in the expected calibration error and Brier score. SIDDA’s versatility across various NN models and datasets, combined with its automated approach to domain alignment, has the potential to significantly advance multi-dataset studies by enabling the development of highly generalizable models.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Machine learning and ligand binding predictions: A review of data, methods, and obstacles

We report that computational predictions of ligand binding is a difficult problem, with more accurate methods being extremely computationally expensive. The use of machine learning for drug binding predictions could possibly leverage the use of biomedical big data in exchange for time-intensive simulations. This paper reviews current trends in the use of machine learning for drug binding predictions, data sources to develop machine learning algorithms, and potential problems that may lead to overfitting and ungeneralizable models. A few popular datasets that can be used to develop virtual high-throughput screening models are characterized using spatial statistics to quantify potential biases. We can see from evaluating some common benchmarks that good performance correlates with models with high-predicted bias scores and models with low bias scores do not have much predictive power. A better understanding of the limits of available data sources and how to fix them will lead to more generalizable models that will lead to novel drug discovery.

59 BASIC BIOLOGICAL SCIENCES↗