Search NASA⌕ Search

SEARCH · Search NASA

Results for “imbalanced data”

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 OSTI2026

HydraGNN_Predictive_GFM_2026 - Ensemble of predictive graph foundation models for atomistic materials modeling

This release contains data and parameters of HydraGNN-based graph foundation models trained as a result of the work published in the pre-print "Exascale Multi-Task Graph Foundation Models for Imbalanced, Multi-Fidelity Atomistic Data" by M. Lupo Pasini et al. (https://arxiv.org/abs/2604.15380). We jointly train on 16 open first-principles datasets (544+ million structures covering 85+ elements) using a multi-task architecture with per-dataset heads and a scalable ADIOS2/DDStore data pipeline. On Frontier, we execute six large-scale DeepHyper hyperparameter optimization campaigns in FP64 and promote the top-performing message-passing models to sustained 2,048-node training, yielding a PaiNN-based lead model. The version of HydraGNN used to generate the outputs provided in this release is HydraGNN v5.0 (https://github.com/ORNL/HydraGNN/releases/tag/v5.0) The list of datasets used for the training of the graph foundation model is the following: 1) Alexandria [1] 2) ANI1x [2] 3) MPTrj [3] 4) Open Catalyst 2020 (OC20) [4] 5) Open Catalyst 2022 (OC22) [5] 6) Open Catalyst 2025 (OC25) [6] 7) Open Direct ir Capture 2023 (ODAC23) [7] 8) Open Materials 2024 (OMat24) [8] 9) Open Molecules 2025 (OMol25) [9] 10) OMol25-neutral (subset of OMol25 that contains only molecules with zero total charge) 11) OMol25-non-neutral (subset of OMol25 that contains only molecules with non-zero total charge) 12) Open Polymers 2026 (OPoly2026) [10] 13) Nabla2DFT [11] 14) QCML [12] 15) QM7X [reference 13] 16) transition1x [14] Dataset references: [1] J. Schmidt et al., “A dataset of 175k stable and metastable materials calculated with the PBEsol and SCAN functionals,” Scientific Data, vol. 9, p. 64, 2022. [2] J. S. Smith et al., “The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules,” Scientific Data, vol. 7, p. 134, 2020. [Online]. Available: https: //www.nature.com/articles/s41597-020-0473-z [3] A. Jain et al., “Commentary: The Materials Project: A materials genome approach to accelerating materials innovation,” APL Materials, vol. 1, no. 1, p. 011002, 07 2013. [Online]. Available: https://doi.org/10.1063/1.4812323 [4] L. Chanussot et al., “Open catalyst 2020 (oc20) dataset and community challenges,” ACS Catalysis, vol. 11, no. 10, pp. 6059–6072, 2021. [Online]. Available: https://doi.org/10.1021/acscatal.0c04525 [5] K. Tran et al., “Open catalyst 2022 (oc22) dataset and challenges for oxidation electrocatalysts,” ACS Catalysis, vol. 13, no. 5, pp. 3066–3084, 2023. [Online]. Available: https://doi.org/10.1021/acscatal.2c05426 [6] S. J. Sahoo et al., “The open catalyst 2025 (oc25) dataset and models for solid-liquid interfaces,” arXiv preprint arXiv:2509.17862, 2025. [Online]. Available: https://arxiv.org/abs/2509.17862 [7] A. Sriram et al., “The open DAC 2023 dataset and challenges for sorbent discovery in direct air capture,” ACS Central Science, vol. 10, no. 5, pp. 923–941, 2024. [8] L. Barroso-Luque et al., “Open materials 2024 (omat24) inorganic materials dataset and models,” 2024. [Online]. Available: https://arxiv.org/abs/2410.12771 [9] D. S. Levine et al., “The open molecules 2025 (OMol25) dataset, evaluations, and models,” 2025. [Online]. Available: https://arxiv.org/abs/2505.08762 [10] D. S. Levine et al., The open polymers 2026 (OPoly26) dataset and evaluations,” arXiv preprint arXiv:2512.23117, 2025. [Online]. Available: https://arxiv.org/abs/2512.23117 [11] K. Khrabrov et al., “Nabla2dft: A universal quantum chemistry dataset of drug-like molecules and a benchmark for neural network potentials,” in NeurIPS 2024 Datasets and Benchmarks Track, 2024. [Online]. Available: https://openreview.net/forum?id=ElUrNM9U8c [12] S. Ganscha et al., “The QCML dataset, quantum chemistry reference data from 33.5M DFT and 14.7B semi-empirical calculations,” Scientific Data, vol. 12, p. 406, 2025. [13] J. Hoja et al., “QM7-X, a comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules,” Scientific Data, vol. 8, p. 43, 2021. [Online]. Available: https://www.nature.com/articles/s41597-021-00812-2 [14] M. Schreiner et al., “Transition1x - a dataset for building generalizable reactive machine learning potentials,” Scientific Data, vol. 9, p. 779, 2022. The folder "datasets_ADIOS2_format" contains the set of pre-processed datasets in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used for the development and training of GFMs in this work. The "datasets_ADIOS2_format" directory contains 2 sub-directories, one for the version "v1" of the datasets and one for the version "v2" of the datasets. The version "v1" of the datasets provides values of the total energy as they are extracted from the original data as it was released by the respective institutions. The version "v2" of the datasets provides values of the energy that have been realigned. The realignment was performed by training a linear regression model that predicts the total energy as a function of the chemical composition of the atomistic structure, and then subtract such prediction from the original value of the total energy. Both folders "v1" and "v2" contain 16 sub-directories, each corresponding to an ADIOS2-formatted dataset The folder "DeepHyper-results" contains the configurational files and model's parameters for all the 186 HPO trials that were successfully completed by the scalable hyperparameter optimization (HPO) runs on Frontier. The content of the folder "DeepHyper-results" I structured as follows: 1) task-list.txt: list of mpnn name, jobid, and deephyper task id 2) gfm_${MPNN}_${JOBID}_0.${TASKID}: run directory with checkpoint files 3) gfm_${MPNN}: deephyper summary directory (*.csv) for each specific MPNN type 4) deephyper-experiment-${JOBID}: output and error logs for each job The file "deephyper-sorted.csv" contains the details of each HydraGNN model built and tested by HPO, obtained by merging the (*.csv) filed from each HPO run executed. Out of all the HPO trials, we selected 10 to continue the training of the respective HydraGNN models. Due to limited computational budget available in the LRN070 allocation we could not complete the training till convergence for all these 10 selected models. The folder "models" contains multiple sub-folders, one per each HydraGNN model trained. Each model sub-folder contains the parameters of each HydraGNN model, with multiple checkpoint-restarts. The list of sub-folders are as follows: 1) multidataset_hpo-BEST1-fp64 2) multidataset_hpo-BEST2-fp64 3) multidataset_hpo-BEST3-fp64 4) multidataset_hpo-BEST4-fp64 5) multidataset_hpo-BEST5-fp64 6) multidataset_hpo-BEST6-fp64 7) multidataset_hpo-BEST7-fp64 8) multidataset_hpo-BEST8-fp64 9) multidataset_hpo-BEST9-fp64 10) multidataset_hpo-BEST10-fp64 Within each one of these folders, additional auxiliary log files are provided with descriptions about how the training proceeded. The lead PaiNN-model is contained inside "multidataset_hpo-BEST6-fp64". The file "mlp_branch_weights" contains the parameters of the multi-layer perceptron (MLP) used to reconcile the predictions of the 16 output decoding heads of the HydragNN architectures. The MLP takes in input the chemical composition of the atomistic structure and predicts averaging weights to linearly mix the predictions of each output decoding head toward consolidating them into a single one. The folder "1.1billion-structure-inference" contains 1.1 billion atomistic structures randomly generated. Each structures is associated with energy and forces predicted with the lead-PaiNN model combined with the MLP model for reconciliation of the multi-branch predictions generated by the 16 output decoding heads. The folder "1.1billion-structure-inference" contains 9,300 (*.tar.gz) subdirectories, one per Frontier compute node used to execute the inference at exascale. Once uncompressed, each (*.tar.gz) subdirectory contains an ADIOS2 (*.bp) file container, where each atomistic structure is stored as a PyTorch-Geometric Data object. The file "export_dataset_environment_variables.sh" contains the environment variables that need to be set before running the HydraGNN code to reproduce the results provided in this dataset release. The code that can be used to load the ADIOS2 files, load HydraGNN models, and run inference is available at: https://github.com/ORNL/HydraGNN/releases/tag/v5.0

36 MATERIALS SCIENCE↗
DOE OSTI2018

Wildfire Mapping in Interior Alaska Using Deep Neural Networks on Imbalanced Datasets

Wildfires are the dominant disturbance impacting many regions in Alaska and is expected to intensify due to climate change. Accurate tracking and quantification of wildfires are important for climate modeling and ecological studies in this region. Remote sensing platforms (e.g., MODIS, Landsat) are an important tool for mapping wildfires in Alaska at varying spatial and temporal scales. Deep neural networks (DNN) have shown superior performance in many classification problems such as high-dimensional remote sensing data. Detection of wildfires is an imbalanced classification problem where one class contains a much smaller or larger sample size and DNNs performance can decline. We take a known weight-selection strategy during DNN training and apply them to MODIS variables (e.g., NDVI, surface reflectance) for a binary classification (i.e., wildfire or no-wildfire) across Alaska during the 2004 wildfire year, which is one of the largest on record. The method splits the input training data into sets, one for training the DNN to update weights and the other that splits into a validation set that is equally balanced between the wildfire and no-wildfire class. The performance is monitored on the validation set to select the weights based on the best validation-loss score. This approach was applied to two sampled datasets, such as where the no-wildfire class can significantly outweigh the wildfire class. The normal DNN training strategy was unable to map wildfires for the highly imbalanced dataset, however, the weight-selection strategy was able to map wildfires very accurately (0.96 recall score for 78,702 wildfire pixels (500×500 m)).

Langford, Zach↗
DOE OSTI2022

Training a Quantum Annealing Based Restricted Boltzmann Machine on Cybersecurity Data

A restricted Boltzmann machine (RBM) is a generative model that could be used in effectively balancing a cybersecurity dataset because the synthetic data a RBM generates follows the probability distribution of the training data. RBM training can be performed using contrastive divergence (CD) and quantum annealing (QA). QA-based RBM training is fundamentally different from CD and requires samples from a quantum computer. We present a real-world application that uses a quantum computer. Specifically, we train a RBM using QA for cybersecurity applications. The D-Wave 2000Q has been used to implement QA. RBMs are trained on the ISCX data, which is a benchmark dataset for cybersecurity. For comparison, RBMs are also trained using CD. CD is a commonly used method for RBM training. Our analysis of the ISCX data shows that the dataset is imbalanced. We present two different schemes to balance the training dataset before feeding it to a classifier. The first scheme is based on the undersampling of benign instances. The imbalanced training dataset is divided into five sub-datasets that are trained separately. A majority voting is then performed to get the result. Our results show the majority vote increases the classification accuracy up from 90.24% to 95.68%, in the case of CD. For the case of QA, the classification accuracy increases from 74.14% to 80.04%. In the second scheme, a RBM is used to generate synthetic data to balance the training dataset. We show that both QA and CD-trained RBM can be used to generate useful synthetic data. Balanced training data is used to evaluate several classifiers. Among the classifiers investigated, K-Nearest Neighbor (KNN) and Neural Network (NN) perform better than other classifiers. They both show an accuracy of 93%. Our results show a proof-of-concept that a QA-based RBM can be trained on a 64-bit binary dataset. The illustrative example suggests the possibility to migrate many practical classification problems to QA-based techniques. Further, we show that synthetic data generated from a RBM can be used to balance the original dataset.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Tree, Shrub, and Grass Classification Using Only RGB Images

In this work, a semantic segmentation-based deep learning method, DeepLabV3+, is applied to classify three vegetation land covers, which are tree, shrub, and grass using only three band color (RGB) images. DeepLabV3+’s detection performance has been studied on low and high resolution datasets that both contain tree, shrub, and grass and some other land cover types. The two datasets are heavily imbalanced where shrub pixels are much fewer than tree and grass pixels. A simple weighting strategy known as median frequency weighting was incorporated into DeepLabV3+ to mitigate the data imbalance issue, which originally used uniform weights. The tree, shrub, grass classification performances are compared when all land cover types are included in the classification and also when classification is limited to the three vegetation classes with both uniform and median frequency weights. Among the three vegetation types, shrub is found to be the most challenging one to classify correctly whereas correct classification accuracy was highest for tree. It is observed that even though the median frequency weighting did not improve the overall accuracy, it resulted in better classification accuracy for the underrepresented classes such as shrub in our case and it also significantly increased the average class accuracy. The classification performance and computation time comparison of DeepLabV3+ with two other pixel-based classification methods on sampled pixels of the three vegetation classes showed that DeepLabV3+ achieves significantly higher accuracy than these methods with a trade-off for longer model training time.

58 GEOSCIENCES↗
DOE OSTI2025

Design Choices in Anomaly Detection for Industrial Control Systems: Insights from Gas Pipeline Data

Industrial control systems (ICS) remain vulnerable to increasingly sophisticated cyberattacks, yet evaluating anomaly detection models in these environments is challenging due to temporal dependencies, missing-not-at-random patterns, and extremely imbalanced datasets. These factors make common practices—especially random data splits and naïve imputation—prone to severe temporal leakage, which can inflate reported performance and obscure real-world limitations. In this work, we systematically examine classical machine learning models, temporal deep learning architecture, and tensor-decomposition–based methods on a gas-pipeline dataset using a fully temporally separated evaluation pipeline designed to mimic realistic deployment conditions. Our findings show that proper temporal handling and MNAR-aware preprocessing significantly alter the relative performance of popular anomaly-detection methods, providing practical guidance for designing reliable, leakage-resistant ICS intrusion-detection systems.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2017

Zoledronate Prevents Simulated Weightlessness-Induced Bone Loss in the Cancellous Compartment While Blunting the Efficacy of a Mechanical Loading Countermeasure

Astronauts using high-force resistance training while weightless show a high-turnover remodeling state within the skeletal system, with resorption and formation biomarkers being elevated. One countermeasure for the skeletal health of astronauts includes an antiresorptive of the bisphosphonate (BP) drug class. We asked, does the combination of an anti-resorptive and high-force exercise during weightlessness have negative effects on bone remodeling and strength? In this study, we developed an integrated model to mimic the mechanical strain of exercise via cyclical loading (CL) in mice treated with the BP Zoledronate (ZOL) combined with hind limb unloading (HU) to simulate weightlessness. We hypothesized that ZOL prevents structural degradation from simulated weightlessness and that CL and ZOL interact to render CL less effective. Thirty-two C57BL/6 mice (male, 16 weeks old, n=8/group) were exposed to 3 weeks of either HU or normal ambulation (NA). Cohorts of mice received one subcutaneous injection of ZOL (45μg/kg), or saline vehicle (VEH), prior to the start of HU. The right tibia was axially compressed in vivo 60x/day to 9N (+1200μstrain on the periosteal surface) and repeated 3x/week during HU. Left tibiae served as a within subject, non-compressed control. Ex vivo μCT was performed on all subjects to determine cancellous and cortical architectural parameters. Static and dynamic histomorphometry were carried out for the left and right tibiae to determine osteoclast- and osteoblast relevant surfaces. Further, micro damage was assessed in select groups by basic-fuchsin staining to test whether CL had an effect. For all assays, a multivariate (2x2x2) ANCOVA model was used to account for body weight changes. Additionally, for the tibiae, we incorporated a random effect for the subject (hence, a mixed model) to account for observations of both left and right tibiae within each subject. P < 0.05 was considered significant. In the cancellous compartment of the proximal tibial metaphysis, we observed a main effect from each independent variable, as determined by structural and histomorphometric assays. Specifically, as expected, ZOL showed an increase in the cancellous bone volume to total volume fraction (BV/TV, +32%) and trabecular number (+18%) compared to the VEH. As expected, ZOL decreased osteoclast surface (OC/BS) by -45% compared to VEH. Surprisingly, ZOL reduced mineralizing surface (MS/BS) and bone formation rate (BFR), indicators of osteoblast activity, by -40% and -54%, respectively, compared to VEH. Altogether, ZOL-treated mice displayed a low turnover state of remodeling in the metaphysis. In the context of skeletal aging, we speculate that ZOL prevented age-related cancellous strut loss during the experiment. As a main effect, as expected, HU decreased BV/TV by - 31% via reductions in both trabecular thickness (-11%) and number (-22%) compared to NA controls. Additionally, HU decreased MS/BS by -38% and bone formation rate (BFR) by -50% compared to NA controls. Altogether, these data are consistent with structural degradation resulting from imbalanced remodeling that favors resorption. As a main effect, CL increased BV/TV by +15% via increased trabecular thickness (+12%) compared to the noncompressed limb. As expected, CL increased MS/BS (+20%) and BFR (+24%), indicating osteoblast mineralization contributed to bone gains. These data show that CL provided an anabolic stimulus to the cancellous tissue. We observed unique interactions in ZOL*CL and HU*CL. First, ZOL prevented CL-induced increases in BV/TV and trabecular number, as compared to VEH. In the context of skeletal aging, these data suggest no added benefit from CL in the ZOL-treated mice. Interestingly, no microdamage was observed in mice that were unloaded and treated with ZOL (independent of CL). Secondly, HU prevented CL-induced increases in BFR, as compared to NA controls. These data suggest that either exercise is less effective or the kinetics of formation are slower during simulated weightlessness. Osteoclast surface was unchanged by either treatment. Thus, in contrast to exercising astronauts, these data do not suggest a high-turnover state in the metaphysis. To assess mechanical properties as a function of HU or ZOL, we tested the left femur in three-point bending ex vivo. As expected, HU decreased stiffness (-30%) compared to NA, and ZOL increased stiffness compared to VEH (+28%). Interestingly, HU increased the post-yield displacement, related to collagenous tensile loading, compared to NA (+20%). ZOL increased yield force (+11%) and ultimate force (+17%), which seems to explain the significant effect of ZOL increasing total energy (work-to-fracture, +15%), while not affecting the post yield displacement. Taken together, ZOL did not have detrimental affect on mechanical properties. Our integrated model simulates the combination of weightlessness, exercise-induced mechanical strain, and anti-resorptive treatment that astronauts experience during space missions. We conclude that Zoledronate was an effective countermeasure against weightlessness-induced bone loss, though zoledronate, as well as weightlessness, rendered exercise-related mechanical loading less effective.

bone↗
DOE OSTI2025

HydraGNN v4.0

The new version of HydraGNN v4.0 provides additional core capabilities, such as: Inclusion of multi-body atomistic cluster expansion MACE, polarizable atom interaction neural network PAINN, and equivariant principal neighborhood aggregation (PNAEq) among the message passing layers supported -Inclusion of graph transformers to directly model long-range interactions between nodes that are distant in the graph topology Integration of graph transformers with message passing layers by combining the graph embedding generated by the two mechanisms, which allows for an improved expressivity of the HydraGNN architecture Improved re-implementation of multi-task learning (MTL) to allow its use for stabilized training across imbalanced, multi-source, multi-fidelity data Introduction of multi-task parallelism, a newly proposed type of model parallelism specifically for MTL architectures, which allows to dispatch different output decoding heads to different GPU devices Integration of multi-task parallelism with pre-existing distributed data parallelism to enable a 2D parallelization for distributed training Improved portability of the distributed training across Intel GPUs, which has been testes on ALCF exascale supercomputer Aurora Inclusion of 2-level fine-grained energy profilers portable across NVIDIA, AMD, and Intel GPUs to monitor the power and energy consumption associated with different functions executed by the HydraGNN code during data pre-load and training Restructuring of previous examples and inclusion of new sets of examples to illustrate the download, preprocess, and training of HydraGNN models on new large-scale open-source datasets for atomistic materials modeling (e.g., Alexandria, Transition1x, OMat24, OMol25)

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗
DOE OSTI2023

FPGA Acceleration of GCN in Light of the Symmetry of Graph Adjacency Matrix

Graph Convolutional Neural Networks (GCNs) are widely used to process large-scale graph data. Different from deep neural networks (DNNs), GCNs are sparse, irregular, and unstructured, posing unique challenges to hardware acceleration with regular processing elements (PEs). In particular, the adjacency matrix of a GCN is extremely sparse, leading to frequent but irregular memory access, low spatial/temporal data locality and poor data reuse. Furthermore, a realistic graph usually consists of unstructured data (e.g., unbalanced distributions), creating significantly different processing times and imbalanced workload for each node in GCN acceleration. To overcome these challenges, we propose an end-to-end hardware-software co-design to accelerate GCNs on resource-constrained FPGAs with the features including: (1) A custom dataflow that leverages symmetry along the diagonal of the adjacency matrix to accelerate feature aggregation for undirected graphs. We utilize either the upper or the lower triangular matrix of the adjacency matrix to perform aggregation in GCN to improve data reuse. (2) Unified compute cores for both aggregation and transform phases, with full support to the symmetry-based dataflow. These cores can be dynamically reconfigured to the systolic mode for transformation or as individual accumulators for aggregation in GCN processing. (3) Preprocessing of the graph in software to rearrange the edges and features to match the custom dataflow. This step improves the regularity in memory access and data reuse in the aggregation phase. Moreover, we quantize the GCN precision from FP32 to INT8 to reduce the memory footprint without losing the inference accuracy. We implement our accelerator design in Intel Stratix10 MX FPGA board with HBM2, and demonstrate 1.3x-110.5x improvement in end-to-end GCN latency as compared to the state-of the-art FPGA implementations, on the graph datasets of Cora, Pubmed, Citeseer and Reddit.

Nair, Gopikrishnan R.↗
DOE OSTI2026

The Impact of Time-Aware Design Choices in ICS Anomaly Detection

Industrial control systems (ICS) remain vulnerable to increasingly sophisticated cyberattacks, yet evaluating anomaly detection models in these environments is challenging due to temporal dependencies, missing-not-at-random patterns, and extremely imbalanced datasets. These factors make common practices—especially random data splits and na¨ıve imputation— prone to severe temporal leakage, which can inflate reported performance and obscure real-world limitations. In this work, we systematically examine classical machine learning models, temporal deep learning architecture, and tensordecomposition– based methods on a gas-pipeline dataset using a fully temporally separated evaluation pipeline designed to mimic realistic deployment conditions. Our findings show that proper temporal handling and MNAR-aware preprocessing significantly alter the relative performance of popular anomaly-detection methods, providing practical guidance for designing reliable, leakage-resistant ICS intrusion-detection systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Results of the Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC)

Abstract Next-generation surveys like the Legacy Survey of Space and Time (LSST) on the Vera C. Rubin Observatory (Rubin) will generate orders of magnitude more discoveries of transients and variable stars than previous surveys. To prepare for this data deluge, we developed the Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC), a competition that aimed to catalyze the development of robust classifiers under LSST-like conditions of a nonrepresentative training set for a large photometric test set of imbalanced classes. Over 1000 teams participated in PLAsTiCC, which was hosted in the Kaggle data science competition platform between 2018 September 28 and 2018 December 17, ultimately identifying three winners in 2019 February. Participants produced classifiers employing a diverse set of machine-learning techniques including hybrid combinations and ensemble averages of a range of approaches, among them boosted decision trees, neural networks, and multilayer perceptrons. The strong performance of the top three classifiers on Type Ia supernovae and kilonovae represent a major improvement over the current state of the art within astronomy. This paper summarizes the most promising methods and evaluates their results in detail, highlighting future directions both for classifier development and simulation needs for a next-generation PLAsTiCC data set.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Structure–activity relationship-based chemical classification of highly imbalanced Tox21 datasets

Abstract The specificity of toxicant-target biomolecule interactions lends to the very imbalanced nature of many toxicity datasets, causing poor performance in Structure–Activity Relationship (SAR)-based chemical classification. Undersampling and oversampling are representative techniques for handling such an imbalance challenge. However, removing inactive chemical compound instances from the majority class using an undersampling technique can result in information loss, whereas increasing active toxicant instances in the minority class by interpolation tends to introduce artificial minority instances that often cross into the majority class space, giving rise to class overlapping and a higher false prediction rate. In this study, in order to improve the prediction accuracy of imbalanced learning, we employed SMOTEENN, a combination of Synthetic Minority Over-sampling Technique (SMOTE) and Edited Nearest Neighbor (ENN) algorithms, to oversample the minority class by creating synthetic samples, followed by cleaning the mislabeled instances. We chose the highly imbalanced Tox21 dataset, which consisted of 12 in vitro bioassays for > 10,000 chemicals that were distributed unevenly between binary classes. With Random Forest (RF) as the base classifier and bagging as the ensemble strategy, we applied four hybrid learning methods, i.e., RF without imbalance handling (RF), RF with Random Undersampling (RUS), RF with SMOTE (SMO), and RF with SMOTEENN (SMN). The performance of the four learning methods was compared using nine evaluation metrics, among which F 1 score, Matthews correlation coefficient and Brier score provided a more consistent assessment of the overall performance across the 12 datasets. The Friedman’s aligned ranks test and the subsequent Bergmann-Hommel post hoc test showed that SMN significantly outperformed the other three methods. We also found that a strong negative correlation existed between the prediction accuracy and the imbalance ratio (IR), which is defined as the number of inactive compounds divided by the number of active compounds. SMN became less effective when IR exceeded a certain threshold (e.g., > 28). The ability to separate the few active compounds from the vast amounts of inactive ones is of great importance in computational toxicology. This work demonstrates that the performance of SAR-based, imbalanced chemical toxicity classification can be significantly improved through the use of data rebalancing.

Idakwo, Gabriel↗
DOE OSTI2022

TULIP: An RNA-seq-based Primary Tumor Type Prediction Tool Using Convolutional Neural Networks

Background: With cancer as one of the leading causes of death worldwide, accurate primary tumor type prediction is critical in identifying genetic factors that can inhibit or slow tumor progression. There have been efforts to categorize primary tumor types with gene expression data using machine learning, and more recently with deep learning, in the last several years. Methods In this paper, we developed four 1-dimensional (1D) Convolutional Neural Network (CNN) models to classify RNA-seq count data as one of 17 highly represented primary tumor types or 32 primary tumor types regardless of imbalanced representation. Additionally, we adapted the models to take as input either all Ensembl genes (60,483) or protein coding genes only (19,758). Unlike previous work, we avoided selection bias by not filtering genes based on expression values. RNA-seq count data expressed as FPKM-UQ of 9,025 and 10,940 samples from The Cancer Genome Atlas (TCGA) were downloaded from the Genomic Data Commons (GDC) corresponding to 17 and 32 primary tumor types respectively for training and validating the models. Results: All 4 1D-CNN models had an overall accuracy of 94.7% to 97.6% on the test dataset. Further evaluation indicates that the models with protein coding genes only as features performed with better accuracy compared to the models with all Ensembl genes for both 17 and 32 primary tumor types. For all models, the accuracy by primary tumor type was above 80% for most primary tumor types. Conclusions: We packaged all 4 models as a Python-based deep learning classification tool called TULIP (TUmor CLassIfication Predictor) for performing quality control on primary tumor samples and characterizing cancer samples of unknown tumor type. Further optimization of the models is needed to improve the accuracy of certain primary tumor types.

Jones, Sara↗
DOE OSTI2018

Filter pruning of Convolutional Neural Networks for text classification: A case study of cancer pathology report comprehension

Convolutional Neural Networks (CNN) have recently demonstrated effective performance in many Natural Language Processing tasks. In this study, we explore a novel approach for pruning a CNN's convolution filters using our new data-driven utility score. We have applied this technique to an information extraction task of classifying a dataset of cancer pathology reports by cancer type, a highly imbalanced dataset. Compared to standard CNN training, our new algorithm resulted in a nearly .07 increase in the micro-averaged F1-score and a strong .22 increase in the macro-averaged F1-score using a model with nearly a third fewer network weights. We show how directly utilizing a network's interpretation of data can result in strong performance gains, particularly with severely imbalanced datasets.

Yoon, Hong-Jun↗
DOE OSTI2022

EE-SMOTE: An oversampling method in conjunction with information entropy for imbalanced learning

Imbalanced learning attracts great attention in various research fields. Existing literature-reported methodologies in imbalanced learning have shown drawbacks including over-generation or noisy/wrong samples generations. This paper presents EE-SMOTE, an oversampling technique based on information entropy, to support the imbalance classifications. Specifically, we propose a metric, Eigen-Entropy (EE), to identify homogenous samples from minority classes for oversampling technique, specifically, SMOTE to reach data balances for classification. Experiments on public dataset and real-world datasets demonstrate the efficacy and effectiveness of the proposed EE-SMOTE in imbalanced learning.

Huang, Jiajing↗
DOE OSTI2021

Asynchronous and Load-Balanced Union-Find for Distributed and Parallel Scientific Data Visualization and Analysis

We present a novel distributed union-find algorithm that features asynchronous parallelism and k-d tree based load balancing for scalable visualization and analysis of scientific data. Applications of union-find include level set extraction and critical point tracking, but distributed union-find can suffer from high synchronization costs and imbalanced workloads across parallel processes. In this study, we prove that global synchronizations in existing distributed union-find can be eliminated without changing final results, allowing overlapped communications and computations for scalable processing. We also use a k-d tree decomposition to redistribute inputs, in order to improve workload balancing. We benchmark the scalability of our algorithm with up to 1,024 processes using both synthetic and application data. Here, we demonstrate the use of our algorithm in critical point tracking and super-level set extraction with high-speed imaging experiments and fusion plasma simulations, respectively.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Deep reinforcement learning for class imbalance fault diagnosis of equipment in nuclear power plants

In equipment fault diagnosis in nuclear power plants, there may be far more samples in one class (e.g., a health state) than in another class (e.g., a fault state). The distribution of data in each class is highly skewed. Most machine learning algorithms are suitable for balanced training datasets. When faced with imbalanced samples, these algorithms tend to provide good identification for the majority classes and bias for the minority classes. However, the misclassification of minority classes can lead to high costs. To address the above problem, this paper develops a deep reinforcement learning-based diagnosis method that models fault diagnosis as a sequential decision-making process. At each time step, the agent receives the state of the environment represented by the training samples and then takes a diagnosis action guided by a policy. If the action is correct/incorrect, the agent receives a positive/negative reward. The reward for minority classes is higher than that for majority classes. The agent’s goal is to obtain as many cumulative rewards as possible in the process, i.e., to identify the sample as correctly as possible. Six demonstration scenarios are constructed, depending on the selected fault datasets and the designed model structures. Experiments show that the proposed method achieves a higher weighted-averaged F1 score than the classical supervised learning method in most cases of class imbalance. Finally, the proposed method has potential applications in the field of class imbalance fault diagnosis of equipment in nuclear power plants.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2023

Towards Improved Decision Making for the Smarter Grid: Geomagnetic Disturbance Implicit Modeling Uncertainty Quantification

Geomagnetic disturbances can greatly disrupt the power grid operation. This is caused by low-frequency changes in the earth’s magnetic field, which upon interaction with the deep earth conductivity induces a changing electric field at the earth’s surface. This in turn induces quasi-direct currents which have the potential to cause protective devices to operate, widespread damage to high-voltage transformers, or voltage collapse due to induced reactive power losses. While these impacts have been known since the 1940s, only recently have standards for GMD response been implemented. Because these standards are heavily based on models, it is important to understand when it is appropriate to make certain modeling assumptions. In this work, implicit modeling uncertainty is quantified for the purpose of enhanced decision making for disturbance impact mitigation. Considering electromagnetic models, modeling assumptions of non-uniform grounding, imbalanced operation, transformer saturation, and geomagnetically induced currents, voltage estimation values are analyzed. Statistical characteristics of modeling uncertainty are obtained by analysis of synthetically generated data. Results statistically characterize geomagnetic disturbances modeling uncertainty, providing an important tool for impact mitigation.

grid analysis↗
DOE OSTI2019

Decoding crystallography from high-resolution electron imaging and diffraction datasets with deep learning

While machine learning has been making enormous strides in many technical areas, it is still massively underused in transmission electron microscopy. To address this, a convolutional neural network model was developed for reliable classification of crystal structures from small numbers of electron images and diffraction patterns with no preferred orientation. Diffraction data containing 571,340 individual crystals divided among seven families, 32 genera, and 230 space groups were used to train the network. Despite the highly imbalanced dataset, the network narrows down the space groups to the top two with over 70% confidence in the worst case and up to 95% in the common cases. As examples, we benchmarked against alloys to two-dimensional materials to cross-validate our deep-learning model against high-resolution transmission electron images and diffraction patterns. We present this result both as a research tool and deep-learning application for diffraction analysis.

97 MATHEMATICS AND COMPUTING↗