Search NASA⌕ Search

SEARCH · Search NASA

Results for “hyperparameter”

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.

106 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

A Scalable Gaussian Process Approach to Shear Mapping with MuyGPs

Analysis of cosmic shear is an integral part of understanding structure growth across cosmic time, which in turn provides us with information about the nature of dark energy. Conventional methods generate shear maps from which we can infer the matter distribution in the universe. Current methods (e.g., Kaiser–Squires inversion) for generating these maps, however, are tricky to implement and can introduce bias. Recent alternatives construct a spatial process prior for the lensing potential, which allows for inference of the convergence and shear parameters given lensing shear measurements. Realizing these spatial processes, however, scales cubically in the number of observations—an unacceptable expense as near-term surveys expect billions of correlated measurements. Therefore, we present a linearly scaling shear map construction alternative using a scalable Gaussian process prior called MuyGPs. MuyGPs avoids cubic scaling by conditioning interpolation on only nearest neighbors and fits hyperparameters using batched leave-one-out cross-validation. This work is the first step toward a full, scalable mass mapping method. We work in a simplified regime where we validate our method by interpolating and analyzing maps given noisy point-estimate data from all three shear fields, taken from a suite of N -body ray-tracing simulations. We also show that we can perform these operations at the scale of billions of galaxies on high-performance computing platforms.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Synthetic Streamflow Datasets Derived from DOE 9505 for Select Texas Basins

This dataset is generated using a Bayesian Hidden Markov Model trained on the DOE 9505 streamflow projection ensemble. A set of 21,000 streamflow realizations are generated for the Colorado, Sabine, and Trinity river basins in Texas. Separate models are trained either using the full 9505 ensemble or a subset based on three hyperparameters: bias correction, downscaling, and hydrological model.

drought↗
DOE OSTI2026

Xanthos-Lake Model Source Code

This repository contains the source code for Xanthos-Lake, a lake-modeling extension of the Xanthos framework that introduces a coupled lake component comprising the Xanthos-Lake Snow and Ice Model (xLSIM) and the Xanthos-Lake Water Balance Model (xLWBM). xLSIM is a basin-aware machine-learning model for lake snow, ice, and thermal conditions. It predicts monthly lake ice thickness, snow depth, snow-cover fraction, mixing-layer temperature, and lake ice fraction from meteorological forcing and lake surface-area information. It uses sequence-based deep-learning architectures, including Transformer and hybrid Long Short-Term Memory–Transformer (LSTM–Transformer) models, together with seasonal encoding, multi-lake learning, physical masking, and basin-level cryospheric and non-cryospheric classification. The training workflow uses Ray for scalable execution and includes optional Ray Tune hyperparameter optimization. Model predictions, observations, diagnostics, and feature-importance outputs are written in NetCDF. xLWBM is the water-balance component of the new lake framework. It simulates monthly lake storage, surface area, evaporation, inflow, outflow, and lake–groundwater exchange. It combines physical water-balance equations with calibrated bathymetric relationships, weir-based outlet flow, modified Penman open-water evaporation, groundwater head relaxation, Penman–Monteith snow and ice sublimation, and snow, ice, and thermal conditions supplied by xLSIM. The model calibrates lake parameters against satellite-derived surface-area data, using evaporation-based calibration where surface-area data are unavailable, and supports small, medium, and large lake classes. For large lakes, xLWBM is integrated with the managed-routing workflow so that lake storage and outflow interact directly with downstream river routing and reservoir operations. Together, xLSIM and xLWBM provide Xanthos with a coupled lake-modeling capability. xLSIM supplies the snow, ice, and thermal conditions that affect lake evaporation and snow- and ice-related water exchanges, while xLWBM translates those conditions into dynamic lake storage, surface area, evaporation, and discharge. In return, xLWBM supplies evolving lake surface area to xLSIM. This coupling enables Xanthos to represent lakes as active hydrologic components within basin-scale water-availability and routing simulations.

Machine Learning↗
DOE OSTI2025

Neural Network Analysis of Nuclear Magnetic Resonance and Infrared Spectra

Nuclear magnetic resonance (NMR) spectroscopy and infrared (IR) spectroscopy are powerful chemical characterization techniques with broad general usage. However, the manual evaluation of the resulting spectra is time-consuming and requires significant expertise, preventing insights from being used in real-time applications. With recent advances in computation and artificial intelligence (AI), new tools are available for automating spectral interpretation. In this work, machine learning (ML) algorithms using 1-dimensional convolutional neural networks (CNNs) were applied to identify common functional groups from spectral information. Raw spectra were collected virtually from the Human Metabolome Database (HMDB) and National Institute of Standards and Technology (NIST) Chemistry WebBook and processed into a suitable standard. Algorithm design was tailored to best fit the nature of the problem, with built-in flexibility to accommodate relevant parameters beyond the raw spectral input, specifically solvent identity and magnetic frequency for NMR. The predictive capability of the algorithm in identifying functional groups is displayed in several examples. This methodology has been compiled into a code repository and could easily be modified to adapt alternative data sources, including other spectrum types. To mitigate overfitting, a common problem in mathematical modeling where overfamiliarity with training data produces trends that are not representative of the general data, a novel metric was developed, referred to as Accufit. Accufit includes a parameter that penalizes substantial differences in the training accuracy and the accuracy of an independent validation set. Examples are presented showing the effectiveness of Accufit in maintaining the model’s predictive capability while controlling the overfitting when used as a custom metric for hyperparameter tuning.

Sturgill, James↗
DOE OSTI2025

GeoLoRA: Geometric integration for parameter efficient fine-tuning

Low-Rank Adaptation (LoRA) has become a widely used method for parameter-efficient fine-tuning of large-scale, pre-trained neural networks. However, LoRA and its extensions face several challenges, including the need for rank adaptivity, robustness, and computational efficiency during the fine-tuning process. We introduce GeoLoRA, a novel approach that addresses these limitations by leveraging dynamical low-rank approximation theory. GeoLoRA requires only a single backpropagation pass over the small-rank adapters, significantly reducing computational cost as compared to similar dynamical low-rank training methods and making it faster than popular baselines such as AdaLoRA. This allows GeoLoRA to efficiently adapt the allocated parameter budget across the model, achieving smaller low-rank adapters compared to heuristic methods like AdaLoRA and LoRA, while maintaining critical convergence, descent, and error-bound theoretical guarantees. The resulting method is not only more efficient but also more robust to varying hyperparameter settings. We demonstrate the effectiveness of GeoLoRA on several state-of-the-art benchmarks, showing that it outperforms existing methods in both accuracy and computational efficiency.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗
DOE OSTI2025

Predicting Initial Trans-Membrane Pressure for Optimized Operations in UF Unit Using Random Forest

With the growing scarcity of freshwater, innovative process design mechanisms like Reverse Osmosis (RO) are increasingly gaining attention among water treatment utilities to address the rising demand. Ensuring reliable water production necessitates efficient resource utilization, minimizing downtime in (ultra-filtration) UF systems. Recent advancements in machine learning (ML) have enabled the development of accurate data-driven models for Model Predictive Control (MPC), often requiring minimal prior knowledge of underlying physical processes. In this study, we present predictive regression models based on Random Forest (RF) and Auto-Regressive (AR) approaches to forecast the initial Trans-Membrane Pressure (TMP) for each filtration cycle in data generated by Direct Potable Reuse (DPR) systems. The proposed RF-based model demonstrates superior performance compared to baseline methods, including historical mean, Last Observation Carried Forward (LOCF), and naïve AR models, across various forecasting horizons in terms of root mean square error (RMSE) metric. To evaluate how different classes of process variables contribute to TMP dynamics over time, we examine the feature importance of independent covariates across multiple forecast horizons. This analysis provides insight into the temporal relevance of operational and sensor-derived features, guiding control and monitoring strategies. Additionally, the impact of hyperparameter tuning on TMP prediction performance is studied for both direct and recursive RF modelling approaches across increasing forecast horizons. Accurate prediction of initial TMP is critical for optimizing RO operations, as it enables the development of robust modelling frameworks by accurately estimating membrane fouling trends, thereby enhancing process efficiency and long-term reliability. The demonstrated efficacy of the RF-based approach highlights its potential as a tool for real-time decision-making in water treatment systems, paving the way for advanced process optimization and sustainable water resource management.

Mukherjee, Subrata [ORNL] (ORCID:0000000309930338)↗
DOE OSTI2025

MoE-Inference-Bench: Performance Evaluation of Mixture of Expert Large Language and Vision Models

Mixture of Experts (MoE) models have enabled the scaling of Large Language Models (LLMs) and Vision Language Models (VLMs) by achieving massive parameter counts while maintaining computational efficiency. However, MoEs introduce several inference-time challenges, including load imbalance across experts and the additional routing computational overhead. To address these challenges and fully harness the benefits of MoE, a systematic evaluation of hardware acceleration techniques is essential. We present MoE-Inference-Bench, a comprehensive study to evaluate MoE performance across diverse scenarios. We analyze the impact of batch size, sequence length, and critical MoE hyperparameters such as FFN dimensions and number of experts on throughput. We evaluate several optimization techniques on Nvidia H100 GPUs, including pruning, Fused MoE operations, speculative decoding, quantization, and various parallelization strategies. Our evaluation includes MoEs from the Mixtral, DeepSeek, OLMoE and Qwen families. The results reveal performance differences across configurations and provide insights for the efficient deployment of MoEs.

Chitty-Venkata, Krishna Teja↗
DOE OSTI2026

Poisson Log-Normal Process for Count Data Prediction

Modeling count data is important in physics and other scientific disciplines, where measurements often involve discrete, non-negative quantities such as photon or neutrino detection events. Traditional parametric approaches can be trained to generate integer-count predictions but may struggle with capturing complex, non-linear dependencies often observed in the data. Gaussian process (GP) regression provides a robust non-parametric alternative to modeling continuous data; however, it cannot generate integer outputs. We propose the Poisson Log-Normal (PoLoN) process, a framework that employs GP to model Poisson log-rates. As in GP regression, our approach relies on the correlations between data points captured via GP kernel structure rather than explicit functional parameterizations. We demonstrate that the PoLoN predictive distribution is Poisson-LogNormal and provide an algorithm for optimizing kernel hyperparameters. Furthermore, we adapt the PoLoN approach to the problem of detecting weak localized signals superimposed on a smoothly varying background - a task of considerable interest in many areas of science and engineering. Our framework allows us to predict the strength, location and width of the detected signals. We evaluate PoLoN's performance using both synthetic and real-world datasets, including the open dataset from CERN which was used to detect the Higgs boson at the Large Hadron Collider. Our results indicate that the PoLoN process can be used as a non-parametric alternative for analyzing, predicting, and extracting signals from integer-valued data.

Saha, Anushka [Rutgers U., Piscataway]↗
DOE OSTI2025

Development of a Discrepancy Checker for the Digital Twin in a Supervisory Control System for a Thermal Energy Delivery System

Defined as a virtual representation of a physical object, process, or service, and used to support real-world decision-making, a digital twin (DT) can be utilized to combine classical and novel frameworks in sensors, state predictions, and multi-input/multi-output systems, and to enable optimal autonomous operations. However, a DT’s usefulness largely depends on its ability to adequately mirror the state of its physical counterpart, and this adequacy should be reflected by the level of uncertainty in the underlying simulation models when estimating and predicting quantities of interest (QOIs). Moreover, simulation models in a DT may involve multiple fidelities of representations—ranging from physics-based models to data-driven ones—but classical uncertainty quantification (UQ) methods struggle to handle numerous uncertainty sources, nor are they designed for real-time applications. This work presents a UQ-based discrepancy checking and diagnosis tool for a DT-based supervisory control system applied to a thermal energy delivery system (TEDS) at Idaho National Laboratory. The discrepancy checker was developed using metadata from an automated DT development process, and these metadata included different combinations of physical model forms and model parameters, training data and hyperparameters for surrogate models, and design parameters for supervisory control systems. Next, correlations between the uncertainty results and the metadata were established and then applied to the DT operations. The discrepancy checker evaluates the discrepancies between model predictions from virtual and sensor measurements and backtraces them to the corresponding major sources of uncertainty. The discrepancy checker showed reasonable performance in detecting discrepancies and diagnosing sources of uncertainty in testing scenarios.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

Investigation of the Performance and Explainability Tradeoffs for Machine-Learning Models for Predictive Maintenance of Circulating Water Systems in Nuclear Power Plants

Predictive maintenance (PdM) has shown great potential for achieving substantial cost savings and enhancing the economic competitiveness of nuclear power plants (NPPs) in today's energy market. Among the different modeling approaches that exist, machine learning (ML) tools in particular have a demonstrated ability to handle high dimensional and multivariate data and to extract hidden relationships within data in industrial environments. While ML methods show great potential, their lack of explainability---especially for black-box models---is a major hurdle to their adoption. Moreover, considering the supposed trade-off between explainability and performance challenges, careful consideration must be made as to which of these quality aspects takes precedence in light of multiple modeling options, resource availability, and domain characteristics. The present work evaluates the performance of six ML models, each with a different degree of explainability, in classifying the conditions of circulating water pumps (CWPs) by utilizing sensor data from nuclear power plants. To determine the drivers behind the trade-offs presented by this array of models, this work also tests different combinations of CWP units as the training and testing data, degrees of data imbalance, and objective functions for hyperparameter tuning. It was found that black-box models tend to afford superior performance in cases where there are far more instances of one type of labeled data than of any other type. It is recommended that a guided procedure be followed for designing and delivering an ML system that is sufficiently explainable to all involved stakeholders.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Neural Scaling Laws for Jet Generation

Recently observed empirical scaling laws describe the performance of foundation-type models as three independent key quantities -- dataset size, compute, and model parameters -- are modified. Extracting these scaling laws informs the training of large complex models for which the tuning of hyperparameters in traditional ways is not feasible. This work for the first time explores if scaling laws can also be observed for the task of particle jet generation -- both relevant as a pre-training objective for foundation models and as in-situ simulation by itself. We indeed replicate the key logarithmic scaling law behavior for model-size scaling. Beyond studying the next token prediction validation loss of the generative model, we also study the sliced Wasserstein distance of five physical quantities that are not immediately available to the model during training. Our study shows that this quantity is monotonically related to the next token prediction validation loss, meaning that this loss is indeed a good proxy for the physics performance. For the scaling with dataset size and compute, we observe substantially weaker scaling behavior of both the loss and the sliced Wasserstein distance. We analyze this behavior by introducing the concept of a learnable window, and argue that autoregressive next token prediction on jet constituents exhibits comparatively rapid saturation relative to language-model studies. We discuss possible origins of this behavior, including the stochastic nature of QCD radiation and differences between generative and supervised learning tasks in collider physics.

Amram, Oz [Fermilab]↗
DOE OSTI2026

Real-Time Detection of Charge Jumps in Superconducting Qubits with a Convolutional Neural Network

Ionizing radiation from cosmic rays and gammas can induce discontinuous jumps in the environmental charge of superconducting qubits (charge jumps), causing correlated errors that challenge fault-tolerant quantum computing while simultaneously providing a detection signature for quantum sensing applications. Current detection methods operate offline, introducing latency incompatible with in-the-loop qubit control. In this paper, an online detector of charge jumps for superconducting qubits, based on a dilated causal convolutional neural network (DCCNN) designed for in-the-loop deployment on the Quantum Instrumentation Control Kit (QICK) platform, is presented. The network is trained on synthetic Ramsey tomography scans generated from qubit templates measured at the Northwestern Experimental Underground Site (NEXUS) at Fermilab, and translated to FPGA firmware via hls4ml with ap_fixed$\langle 16,6 \rangle$ quantization, reaching a per-inference latency of $6.19 μ$s on the Zynq UltraScale+ RFSoC ZCU216. At this operating point the DCCNN matches the detection efficiency of the established offline $χ^2$ algorithm ($0.843 \pm 0.022$ vs. $0.866 \pm 0.020$ on $|Δq| \in [0.1, 0.5] e$ at matched false-positive rate), while requiring no per-qubit hyperparameter tuning. This shifts charge-jump detection from a post-hoc diagnostic to a control-loop primitive, enabling adaptive protocols that respond to radiation-induced events in situ, with applications to quantum-computing error mitigation and to the use of superconducting qubits as particle detectors.

Gaytan-Villarreal, Daniel [Carnegie Mellon U.]↗
DOE OSTI2026

HARMONY: Large-Scale Architecture Search for Efficient Hybrid Language Models

As large language models scale to trillions of parameters, their computational and memory requirements present critical challenges for efficient training and deployment. While Mixture of Experts (MoE) architectures enable efficient scaling through sparse parameter activation, and state-space models like Mamba offer linear-time complexity, principled methods for combining these paradigms remain undeveloped. We introduce HARMONY (Hybrid Architecture Research for Mamba, Optimized with Neural efficiencY), a multi-objective evolutionary neural architecture search framework for discovering efficient hybrid language models that integrate Transformer attention mechanisms, Mixture-of-Experts routing, and Mamba state-space components. Through large-scale distributed search using 16,384 MI250X GPUs on the Frontier supercomputer, HARMONY explores a comprehensive design space encompassing six attention variants (MHA, MQA, GQA, MLA, SWA, and Mamba-2), variable MoE configurations with both routed and shared experts, and extensive Mamba hyperparameters. Our framework discovers heterogeneous architectures that balance training performance with computational efficiency through multi-objective optimization incorporating latency penalties and fitness-based selection. Analysis of discovered architectures reveals that optimal hybrid designs favor heterogeneous component mixing rather than homogeneous patterns, with Mamba-2 and Multi-Head Latent Attention (MLA) emerging as preferred mechanisms. Discovered architectures demonstrate superior training efficiency: our best configuration achieves a final perplexity of 1.0874 with 2.38B parameters while processing 4,320 tokens/second, outperforming significantly larger manually designed models. Full-scale evaluation shows HARMONY's top architectures achieve better loss trajectories than equivalently-sized models using state-of-the-art configurations including Mixtral, Jamba, and Samba. Additionally, we demonstrate 91% weak scaling efficiency when training discovered 36B-parameter models across 1,024 GPUs. HARMONY is released as an open framework with comprehensive tools for building and training hybrid models using expert-data-pipeline parallelism, democratizing access to automated architecture design for next-generation language models.

Herron, Emily [ORNL] (ORCID:0000000273008172)↗
DOE OSTI2026

Fisher Forecasting for the DESC with $\texttt{Augur}$

The Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) has begun its ten-year survey of the entire visible southern hemisphere. To ensure robust cosmological measurements, computationally inexpensive investigations of modeling choices must be made to gauge the performance of proposed cosmological analyses. In this paper, we introduce the $\texttt{Augur}$ tool of the Dark Energy Science Collaboration (DESC), which provides Fisher forecasts for cosmological inference for the LSST using software frameworks designed for DESC science. We test the pipeline by comparing it to forecasts produced by external code and direct sampling of the posterior via nested sampling methods, finding good agreement between all methods. We additionally investigate a range of modeling and hyperparameter choices for a 3$\times$2pt investigation in harmonic space, providing users with diagnostics to obtain reliable forecasts. $\texttt{Augur}$ will be continually updated to be compatible with the other tools in the DESC software ecosystem as additional probes and functionality become available.

Rogozenski, Paul [Carnegie Mellon U.; Arizona U.] ↗
DOE OSTI2024

Bayes_Opt-SWMM: A Gaussian process-based Bayesian optimization tool for real-time flood modeling with SWMM

Real-time flood model plays a pivotal role in averting urban flood damage, particularly when there is minimal lead time for preparatory measures. However, urban flood modeling in real-time often contends with inherent uncertainties arising from input data uncertainty and parameter ambiguities. Here this study introduces a real-time calibration (RTC) tool called Bayes_Opt-SWMM, specifically tailored for real-time urban flood modeling and uncertainty optimization. This tool leverages the Gaussian process-based Bayesian optimization algorithm and interfaces seamlessly with the Stormwater Management Model (SWMM). It integrates real-time model forcing data and flood monitoring collected through sensors and gauges which are strategically placed within critical locations of urban drainage systems. Our approach hinges on the Surrogate Model based Uncertainty Optimization (SMUO) concept, providing an avenue for enhancing real-time flood modeling. Bayes_Opt-SWMM runs the optimization process using a surrogate model called Gaussian Process emulator with two inference methods: (1) the Gaussian Process (GP) model and (2) Markov Chain Monte Carlo (MCMC) algorithm in GP model (GP_MCMC). Furthermore, three acquisition functions, namely Expected Improvement (EI), Maximum Probability of Improvement (MPI), and Lower Confidence Bound (LCB), facilitate optimal parameter fitting within the surrogate models. The efficiency of GP-based surrogate models in learning SWMM model parameters, leads to an improved uncertainty quantification and accelerated real-time flood modeling in urban areas. Overall, Bayes_Opt-SWMM emerges as a cost-effective and valuable tool for real-time flood modeling and monitoring, with significant potential for managing intelligent storm water systems in urban environments.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Benchmarking the performance of uncertainty quantification methods for neural network-based interatomic potentials

Machine-learned interatomic potentials (ML-IAPs) continue to gain popularity as accurate, computationally efficient replacements for traditional, physics-based interatomic potentials and expensive ab initio methods. Uncertainty quantification (UQ) of ML-IAPs is a growing area of research as UQ is critical in many applications of IAPs, such as developing curated datasets, active learning-based data augmentation, self-improving models, and estimating the uncertainty of molecular dynamics simulations. In this paper, we construct and benchmark a series of different neural network potentials (NNPs) with varying network architectures to determine the performance of these models with respect to both the mean and uncertainty calibration error. Each NNP method is specifically designed to predict either epistemic or aleatoric uncertainty with particular focus on the differences in behavior between the epistemic and aleatoric uncertainty estimates. We benchmark these methods using multiple datasets common in the ML-IAP literature. The results show that the aleatoric uncertainty from single-shot model architectures is a competitive alternative to ensemble-based epistemic uncertainty predictions in regions of sufficient data-density. However, in regions where the representative data is sparse, aleatoric uncertainty models tend to overpredict and epistemic methods tend to underpredict the actual model error. We conclude that the type of UQ is crucial when discussing performance of probabilistic model results as different methods have different performance characteristics depending on the regime in which they are evaluated. Therefore, the type of UQ method should be carefully evaluated against both the data characteristics and requirements for the intended application.

97 MATHEMATICS AND COMPUTING↗