Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “evaluation datasets”

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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
NASA NTRS2022

Development of a “Nature Run” for Observing System Simulation Experiments (OSSEs) for Snow Mission Development

Snow is a fundamental component of global and regional water budgets, particularly in mountainous areas and regions downstream that rely on snowmelt for water resources. Land surface models (LSMs) are commonly used to develop spatially distributed estimates of snow water equivalent (SWE) and runoff. However, LSMs are limited by uncertainties in model physics and parameters, among other factors. In this study, we describe the use of model calibration tools to improve snow simulations within the Noah-MP LSM as the first step in an observing system simulation experiment (OSSE). Noah-MP is calibrated against the University of Arizona (UA) SWE product over a western Colorado domain. With spatially varying calibrated parameters, we run calibrated and default Noah-MP simulations for water years 2010–20. By evaluating both simulations against the UA dataset, we show that calibration decreases domain averaged temporal RMSE and bias for snow depth from 0.15 to 0.13 m and from −0.036 to −0.0023 m, respectively, and improves the timing of snow ablation. Increased snow simulation performance also improves estimates of model-simulated runoff in four of six study basins, though only one has statistically significant improvement. Spatially distributed Noah-MP snow parameters perform better than default uniform values. We demonstrate that calibrating variables related to snow albedo calculations and rain–snow partitioning, among other processes, is a necessary step for creating a nature run that reasonably approximates true snow conditions for the OSSEs. Additionally, the inclusion of a snowfall scaling term can address biases in precipitation from meteorological forcing datasets, further improving the utility of LSMs for generating reliable spatiotemporal estimates of snow.

Melissa L. Wrzesien↗
NASA NTRS2022

Cloud Mask Intercomparison eXercise (CMIX): An evaluation of cloud masking algorithms for Landsat 8 and Sentinel-2

Cloud cover is a major limiting factor in exploiting time-series data acquired by optical spaceborne remote sensing sensors. Multiple methods have been developed to address the problem of cloud detection in satellite imagery and a number of cloud masking algorithms have been developed for optical sensors but very few studies have carried out quantitative intercomparison of state-of-the-art methods in this domain. This paper summarizes results of the first Cloud Masking Intercomparison eXercise (CMIX) conducted within the Committee Earth Observation Satellites (CEOS) Working Group on Calibration & Validation (WGCV). CEOS is the forum for space agency coordination and cooperation on Earth observations, with activities organized under working groups. CMIX, as one such activity, is an international collaborative effort aimed at intercomparing cloud detection algorithms for moderate-spatial resolution (10–30 m) spaceborne optical sensors. The focus of CMIX is on open and free imagery acquired by the Landsat 8 (NASA/USGS) and Sentinel-2 (ESA) missions. Ten algorithms developed by nine teams from fourteen different organizations representing universities, research centers and industry, as well as space agencies (CNES, ESA, DLR, and NASA), are evaluated within the CMIX. Those algorithms vary in their approach and concepts utilized which were based on various spectral properties, spatial and temporal features, as well as machine learning methods. Algorithm outputs are evaluated against existing reference cloud mask datasets. Those datasets vary in sampling methods, geographical distribution, sample unit (points, polygons, full image labels), and generation approaches (experts, machine learning, sky images). Overall, the performance of algorithms varied depending on the reference dataset, which can be attributed to differences in how the reference datasets were produced. The algorithms were in good agreement for thick cloud detection, which were opaque and had lower uncertainties in their identification, in contrast to thin/semi-transparent clouds detection. Not only did CMIX allow identification of strengths and weaknesses of existing algorithms and potential areas of improvements, but also the problems associated with the existing reference datasets. The paper concludes with recommendations on generating new reference datasets, metrics, and an analysis framework to be further exploited and additional input datasets to be considered by future CMIX activities.

Sergii Skakun↗
NASA NTRS2022

Design of Materials with Alchemite

Machine learning models that establish the relationships between materials processing and properties can enable inverse design of materials through active learning. Alchemite is a commercial software that can perform inverse materials design on sparse data. Here we evaluate Alchemite’s performance on a dataset of shape memory alloys and a dataset of heat exchangers compared to baseline random forest models. Alchemite had higher accuracy when making predictions on sparse data and was more accurate or nearly as accurate as random forests on complete datasets while also quantifying uncertainty. The software was also used to suggest processing steps and design parameters to optimize properties and performance; however, physical validation of the suggested design parameters was beyond the scope of this work. Several useful design insights were gained about the impact of the design parameters on properties and performance including the importance of dopant choice and amount for shape memory alloys and the importance of height and weight on the thermal resistance of heat exchangers.

Machine learning↗
NASA NTRSDate not supplied

Classification of Notices to Airmen using Natural Language Processing

This paper establishes the feasibility of using Natural Language Processing (NLP) to classify NOTAMs or Notices to Airmen – a pilot messaging framework to gather real-time situational awareness. Present day air mobility operations heavily rely on NOTAMs. However, pilots often have difficulty interpreting NOTAMs due to the sheer volume of inapplicable messages and unclear abbreviations. Using NLP, the presented study analyzes the accuracy of classifying NOTAMs and, thereby, the efficiency of generating actionable interpretations in real time. To this effect, efficacies of four NLP neural network architectures were analyzed, including three Recurrent Neural Networks (RNNs) with GloVe, Word2Vec, and FastText word embeddings, and one trained Bi-Directional Encoder Representations from Transformers (BERT) model. The four neural networks were trained and evaluated on three open-source datasets of varying text lengths, vocabularies, and grammars, taken from e-commerce product descriptions, social media tweets, and unstructured descriptions for data and analytics services on open data marketplaces such as NASA’s Data and Reasoning Fabric (DRF) platform. This provided cross-analysis of each neural network architecture’s performance per text type. The best performing architecture, BERT, was then fine-tuned on a collection of open-source NOTAM data. Post-training, a real-time NOTAM classification service was implemented to draw inference on new NOTAMs using the trained model, which demonstrated close to 99% accuracy in classification. This modular classification service is envisioned to be integrated with a data and analytics delivery platform, such as the DRF, thus availing real-time contextualization of NOTAMs to air mobility clients, humans, and machines for enhanced decision making.

Aiden C. Szeto↗
NASA NTRSDate not supplied

Classification of Notices to Airmen using Natural Language Processing

This paper establishes the feasibility of using Natural Language Processing (NLP) to classify NOTAMs or Notices to Airmen – a pilot messaging framework to gather real-time situational awareness. Present day air mobility operations heavily rely on NOTAMs. However, pilots often have difficulty interpreting NOTAMs due to the sheer volume of inapplicable messages and unclear abbreviations. Using NLP, the presented study analyzes the accuracy of classifying NOTAMs and, thereby, the efficiency of generating actionable interpretations in real time. To this effect, efficacies of four NLP neural network architectures were analyzed, including three Recurrent Neural Networks (RNNs) with GloVe, Word2Vec, and FastText word embeddings, and one trained Bi-Directional Encoder Representations from Transformers (BERT) model. The four neural networks were trained and evaluated on three open-source datasets of varying text lengths, vocabularies, and grammars, taken from e-commerce product descriptions, social media tweets, and unstructured descriptions for data and analytics services on open data marketplaces such as NASA’s Data and Reasoning Fabric (DRF) platform. This provided cross-analysis of each neural network architecture’s performance per text type. The best performing architecture, BERT, was then fine-tuned on a collection of open-source NOTAM data. Post-training, a real-time NOTAM classification service was implemented to draw inference on new NOTAMs using the trained model, which demonstrated close to 99% accuracy in classification. This modular classification service is envisioned to be integrated with a data and analytics delivery platform, such as the DRF, thus availing real-time contextualization of NOTAMs to air mobility clients, humans, and machines for enhanced decision making.

Aiden Szeto↗
DOE OSTI2024

Spatiotemporal Automatic Calibration of Infrastructure Lidar, Radar, and Camera with a Global Navigation Satellite System: Preprint

Robust and accurate perception is important for modern intelligent transportation systems (ITS), which use sensors of various modalities for data fusion to create a digital twin of an intersection. Sensor calibration is an important process that creates a unified coordinate frame for the sensor output data so that it can be used for data fusion. Classical approaches for sensor calibration are time-consuming, require an overlapping field of view for feature matching, and are not feasible for ITS application as they cause disruptions in the flow of traffic. In this paper, we present a spatiotemporal automatic calibration approach to calibrate multiple infrastructure lidar, radar, and cameras installed at a traffic intersection. The approach uses global navigation satellite system (GNSS) positioning information shared by connected vehicles, and when the vehicle is detected by the sensor, we match the sensor detections with the GNSS coordinates. The proposed algorithm is evaluated with a real-world dataset utilizing detections from two radars, cameras, and lidars with a test vehicle instrumented with a post-processing kinematic (PPK)-corrected GNSS driving past the sensors installed at a four-way traffic intersection. The experimental results show that the proposed automatic calibration approach can achieve the transformation with a root mean squared error of less than 0.5 for radar and lidar and less than 2 for camera detections. The ability to rapidly calibrate sensors not only benefits initial installations, but can also be used for system health monitoring, while utilizing available connected vehicle data to test the real-time sensor fidelity and operational status.

ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2025

Review of Grey Box/Black Box Data Contamination Metrics on Open and Commercial Models

Dataset contamination is a problem where benchmarks and tasks used to evaluate the capabilities of Large Language Models (LLMs) have been incorporated into the training dataset of the models. This gives a false sense of performance that can overestimate how these models will function on truly unseen data. This problem becomes worse with commercial LLMs with larger and non-accessible training data, so techniques have been developed to try to measure the degree to which a model is contaminated with a benchmark’s data. To understand the effectiveness of these techniques, particularly when evaluating contamination on coding tasks, we review trends and categorize techniques by the degree of access to the model that is required. The research literature on this topic has reported mixed effectiveness of these techniques, so we select a set of black box (text access only) and grey box (access to model loss/probabilities required) techniques and apply them to both commercial and non-commercial models. We implement these metrics as part of a framework to test the contamination of Python code in LLMs to see to what extent we can replicate the effectiveness (or ineffectiveness) of these contamination detection techniques. Though we find mixed results in the capabilities of these metrics to identify contamination, we do observe evidence that they can identify contamination (broadly) in fine-tuned models when both a baseline and fine-tuned model is present. Additionally, similarity metrics were able to identify between contaminated and uncontaminated data even in situations where the data is distributionally similar (e.g., drawn from the same set of code projects).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Hydropower Energy Storage Capacity (HESC) Dataset

The Hydropower Energy Storage Capacity (HESC) Dataset catalogs characteristics that are relevant to evaluating reservoir storage and estimates of energy storage capacity based on varying levels of detail. Hydropower dams and reservoirs were included based on information from the National Inventory of Dams (NID; USACE, 2021) and Global Reservoir and Dam (GRanD v1.3) and Existing Hydropower Assets datasets. These data provide a foundation for understanding available resources at existing hydropower facilities and their potential to provide storage of energy and more flexible generation. Estimates of energy storage capacity include: • Level 1 – nominal energy storage capacity based on maximum storage capacities and hydraulic head • Level 2 – nominal energy storage capacity based on historical models or observations of reservoir volume and hydraulic head. These estimates are provided based on capacity from the entire historical period as well as monthly values. • Level 3 – modeled energy generation based on volume-elevation relationships, historical storage, observed/modeled inflows, and hydraulic capacity of turbines and calculated both as overall and on a monthly basis. • Level 4 – modeled energy generation incorporating information from Level 3 and operational constraints. For facilities where installed capacity is known, there are also estimates for discharge duration (the length of time when a facility could provide generation at a given capacity).

13 HYDRO ENERGY↗
NASA NTRS2018

The SUMup Dataset: Compiled Measurements of Surface Mass Balance Components over Ice Sheets and Sea Ice with Analysis over Greenland

Increasing atmospheric temperatures over ice cover affect surface processes, including melt, snowfall, and snow density. Here, we present the Surface Mass Balance and Snow on Sea Ice Working Group (SUMup) dataset, a standardized dataset of Arctic and Antarctic observations of surface mass balance components. The July 2018 SUMup dataset consists of three subdatasets, snow/firn density (https://doi.org/10.18739/A2JH3D23R), at least near-annually resolved snow accumulation on land ice (https://doi.org/10.18739/A2DR2P790), and snow depth on sea ice (https://doi.org/10.18739/A2WS8HK6X), to monitor change and improve estimates of surface mass balance. The measurements in this dataset were compiled from field notes, papers, technical reports, and digital files. SUMup is a compiled, community-based dataset that can be and has been used to evaluate modeling efforts and remote sensing retrievals. Active submission of new or past measurements is encouraged. Analysis of the dataset shows that Greenland Ice Sheet density measurements in the top 1m do not show a strong relationship with annual temperature. At Summit Station, Greenland, accumulation and surface density measurements vary seasonally with lower values during summer months. The SUMup dataset is a dynamic, living dataset that will be updated and expanded for community use as new measurements are taken and new processes are discovered and quantified.

Montgomery, Lynn↗
NASA NTRS2011

Evaluation of the Event Driven Phenology Model Coupled with the VegET Evapotranspiration Model Through Comparisons with Reference Datasets in a Spatially Explicit Manner

A new model coupling scheme with remote sensing data assimilation was developed for estimation of daily actual evapotranspiration (ET). The scheme represents a mix of the VegET, a physically based model to estimate ET from a water balance, and an event driven phenology model (EDPM), where the EDPM is an empirically derived crop specific model capable of producing seasonal trajectories of canopy attributes. In this experiment, the scheme was deployed in a spatially explicit manner within the croplands of the Northern Great Plains. The evaluation was carried out using 2007-2009 land surface forcing data from the North American Land Data Assimilation System (NLDAS) and crop maps derived from remotely sensed data of NASA's Moderate Resolution Imaging Spectroradiometer (MODIS). We compared the canopy parameters produced by the phenology model with normalized difference vegetation index (NDVI) data derived from the MODIS nadir bi-directional reflectance distribution function (BRDF) adjusted reflectance (NBAR) product. The expectations of the EDPM performance in prognostic mode were met, producing determination coefficient (r2) of 0.8 +/-.0.15. Model estimates of NDVI yielded root mean square error (RMSE) of 0.1 +/-.0.035 for the entire study area. Retrospective correction of canopy dynamics with MODIS NDVI brought the errors down to just below 10% of observed data range. The ET estimates produced by the coupled scheme were compared with ones from the MODIS land product suite. The expected r2=0.7 +/-.15 and RMSE = 11.2 +/-.4 mm per 8 days were met and even exceeded by the coupling scheme0 functioning in both prognostic and retrospective modes. Minor setbacks of the EDPM and VegET performance (r2 about 0.5 and additional 30 % of RMSR) were found on the peripheries of the study area and attributed to the insufficient EDPM training and to spatially varying accuracy of crop maps. Overall the experiment provided sufficient evidence of soundness and robustness of the EDPM and VegET coupling scheme, assuring its potential for spatially explicit applications.

Kovalskyy, V.↗
NASA NTRS2000

Visual detection of spatial contrast patterns: evaluation of five simple models

The ModelFest Phase One dataset is a collection of luminance contrast thresholds for 43 two-dimensional monochromatic spatial patterns confined to an area of approximately two by two degrees. These data were collected by a collaboration among twelve laboratories, and were designed to provide a common database for calibration and testing of spatial vision models. Here I report fits of the ModelFest data with five models: Peak Contrast, Contrast Energy, Generalized Energy, a Gabor Channels model, and a Discrete Cosine Transform model. The Gabor Channels model provides the best fit, though the other, simpler models, with the exception of Peak Contrast, provide remarkably good fits as well. Though there are clear individual differences, regularities in the data suggest the possibility of constructing a standard observer for spatial vision. c2000 Optical Society of America.

NASA Discipline Space Human Factors↗
DOE OSTI2024

RODEM Jet Datasets

We present the RODEM Jet Datasets, a comprehensive collection of simulated large-radius jets designed to support the development and evaluation of machine-learning algorithms in particle physics. These datasets encompass a diverse range of jet sources, including quark/gluon jets, jets from the decay of W bosons, top quarks, and heavy new-physics particles. The datasets provide detailed substructure information, including jet kinematics, constituent kinematics, and track displacement details, enabling a wide range of applications in jet tagging, anomaly detection, and generative modelling.

FOS: Physical sciences↗
DOE OSTI2021

Adoption of image-driven machine learning for microstructure characterization and materials design: A Perspective

Microstructure characterization enables the development of structure-processing-property relationships critical to several research areas within the broad field of materials science, from alloy design to the assessment of corrosion resistance, and failure analysis. Conventional approaches to material characterization have relied on either qualitative inference by the human ex-pert or software applications that can extract high-level features from images, such as boundary segmentation, average grain diameter, etc. Such approaches rely heavily on subject matter expert user intervention and knowledge of what phases or more generally, what microstructural features, are of interest. The recent surge in the adoption of machine learning techniques to address problems in materials engineering has brought with it an increased interest and application of Image Driven Machine Learning (IDML) approaches. In this work, we review the applications of IDML to the field of materials characterization. A canonical hierarchy of stages is defined, which when put sequentially together completes an IDML study: problem definition, dataset building, model selection and training, model evaluation, and integration with existing instrumentation or simulation workflow. The studies reviewed in this work are analyzed from the perspective of each of these stages. Such a review permits agranular assessment of the field, for example the impact of IDML on materials characterization at the nanoscale, the size of a typical dataset required to train a semantic segmentation model on electron microscopy images, ubiquitousness of transfer learning in the domain, etc. Finally, we discuss the importance of interpretability and explainability in the field of IDML for materials characterization, and provide an overview of two emerging techniques in the field: semantic segmentation and generative adversarial networks.

Baskaran, Arun↗
DOE OSTI2025

Performance of wind assessment datasets in United States coastal areas

The atmospheric dynamics that occur near the intersection of land and water offer exciting and challenging opportunities for wind energy deployment in coastal locations. New models and tools are continually being developed in support of wind resource assessment, and three recent products are explored in this work for their performance in representing characteristics of the wind resource at coastal locations: the Global Wind Atlas 3 (GWA3), the 2023 National Offshore Wind dataset (NOW-23), and the wind climate simulations that are a component of the Wind Integration National Dataset (WIND) Toolkit Long-Term Ensemble Dataset (WTK-LED Climate). These relatively new products are freely available and user-friendly so that anyone – from a utility-scale developer to a resident or business owner – can evaluate the potential for wind energy generation at their location of interest. The validations in this work provide guidance on the accuracy of wind resource assessments for coastal customers interested in installing small or midsize wind turbines (≤ 1 MW in capacity) to support energy needs at the residential, business, or community scale, such as the island and remotely located participants of the U.S. Department of Energy's Energy Transitions Initiative Partnership Project. At 23 coastal locations across the United States, dataset performance varies according to different evaluation metrics. All three recent datasets tend to overestimate the observed coastal wind resource. GWA3 produces the smallest annual average wind speed relative errors, whereas WTK-LED Climate is in best agreement in terms of representing diurnal wind speed cycles. NOW-23 is the highest performing of the datasets for representing seasonal and interannual trends in the coastal wind resource. While GWA3 and WTK-LED Climate are relatively insensitive to the dataset output heights selected for wind resource assessment at small and midsize wind turbine hub heights (20–60 m), significant variation in the NOW-23 representation of wind shear across the wind profile in the lowest 100 m of the atmosphere leads to notable differences in wind speed estimates according to the dataset output heights selected for evaluation. GWA3 exhibits challenges in the representation of observed wind speed diurnal cycles at small and midsize turbine hub heights, likely due to the dataset's consistent treatment of hourly wind speed trends regardless of altitude.

17 WIND ENERGY↗
DOE OSTI2022

Role of copper in nitrous oxide accumulation in wetland soils from Illinois and South Carolina and stream sediments from Tennessee

Denitrification is microbially-mediated through enzymes containing metal cofactors. Laboratory studies of pure cultures have highlighted that the availability of copper (Cu), required for the multicopper enzyme nitrous oxide reductase, can limit nitrous oxide (N2O) reduction. However, in natural aquatic systems, such as wetlands and hyporheic zones in stream beds, the role of Cu in controlling denitrification remains incompletely understood. In this study, we collected soils and sediments from three natural environments -- riparian wetlands, marsh wetlands, and a stream -- to investigate their nitrogen species transformation activity at background Cu levels and different supplemented Cu loadings. All of the systems contained solid-phase associated Cu below or around geological levels (40–280 nmol g-1) and exhibited low dissolved Cu (3–50 nM), which made them appropriate sites for evaluating the effect of limited Cu availability on denitrification.The dataset contains the variation in the nutrient (nitrate (NO3-), nitrite (NO2-), and ammonium (NH4+)) and N2O concentrations during the incubation experiments to evaluate the effect of Cu. We have also reported the variation in metal concentrations (Cu, iron(Fe) and manganese (Mn)) and dissolved organic carbon (DOC) concentrations during the incubations. The parameters obtained using a kinetic model to quantitatively report the effect of Cu on nitrogen species conversion in natural aquatic systems are also added in the dataset. The dataset also contains the labile concentration of Cu estimated using a speciation model in MINTEQ. All data are provided in text-based CSV format with header sections indicating the data contained in each file and the corresponding units.Our study suggests that high concentrations of N2O accumulated in all microcosms lacking Cu amendment except for one stream sediment sample. With Cu added to provide dissolved concentrations at trace levels (10–300 nM), the reduction rate of N2O to N2 in the wetland soils and stream sediments was enhanced. A kinetic model could account for the trends in nitrogen species by combining the reactions for microbial reduction of NO3- to NO2-/N2O/N2 and abiotic reduction of NO2 to nitrogen (N2_. The model revealed that the rate of N2O to N2 conversion increased significantly in the presence of Cu. For riparian wetland soils and stream sediments, the kinetic model also suggested that overall denitrification is driven by abiotic reduction of NO2- in the presence of inorganic electron donors. This study demonstrated that natural aquatic systems containing Cu at concentrations less than or equal to crustal abundances may display incomplete reduction of N2O to N2 that would cause N2O accumulation and release to the atmosphere.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Gulf of Mexico Miocene CO2 Site Characterization Mega Transect

The project Gulf of Mexico Miocene CO2 Site Characterization Mega Transect is one of 9 site characterization projects that were implemented as part of ARRA (American Recovery and Reinvestment Act). Data from this project was used to improve resolution of data in NATCARB in the area of study. Data related to this study has already been incorporated in NATCARB Atlas. The University of Texas at Austin study focused on (1) collecting high-resolution, 3-D (HR3D) seismic datasets of the shallow geologic section to evaluate the capacity of potential reservoirs and the sealing capacity of the confining system, and (2) producing a regional CO2 atlas of Miocene age units of the upper Texas coast submerged lands (southeast Texas). The goal of the study was to evaluate the capability of the Miocene age geologic section of Texas submerged lands to permanently store large volumes of anthropogenic CO2. The geographic locations of three HR3D, “P-Cable” seismic surveys collected in the near-shore waters of southeast Texas offers a vertical and horizontal view (data “cube”) example from the second survey. The dataset's high quality allowed for defining recognizable geologic morphologies (shapes) and possible fluid content, which are important for determining an area’s prospects for geologic storage. The regional CO2 atlas of Miocene age units highlights several topics, including: (1) a regional analyses of petroleum systems as analogs for CO2 storage; (2) petrography, petrology, and extent of confining systems of the Miocene section; (3) static capacity estimate; and (4) examples of characterization methodologies of prospect areas. The HR3D datasets and regional analyses within the CO2 atlas of the offshore Texas Miocene provide a sound basis for a future generation of specific CO2 prospects in the study area.

.las↗
DOE OSTI2023

Probing Accuracy-Speedup Tradeoff in Machine Learning Surrogates for Molecular Dynamics Simulations

The performance promise of machine learning surrogates of molecular dynamics simulations of soft materials is significant but generally comes at the cost of acquiring large training datasets to learn the complex relationships between input soft material attributes and output properties. Under the constraint of limited high-performance computing resources, optimizing the size of the training datasets becomes paramount. Using an artificial neural network based surrogate for molecular dynamics simulations of confined electrolytes, we explore the tradeoff between surrogate accuracy and computational gains. Accuracy is assessed by computing the root-mean-square errors between the surrogate predictions and the ground truth results obtained via molecular dynamics simulations. The computational performance is judged by evaluating the speedup which incorporates the training dataset creation time. Improvement in accuracy occurs with a loss of speedup, which scales as the inverse of the training dataset size. Furthermore, the link between surrogate generalizability and the accuracy-speedup tradeoff is assessed by examining the errors incurred in surrogate predictions on unseen, interpolated input variables and developing a net speedup metric to capture the associated gains.

Anions↗