Engineering Papers⌕ Search

SEARCH · Engineering Papers

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.

At least 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Replicating Machine Learning Experiments in Materials Science

Transparency and reproducibility are important aspects of validation for Machine Learning (ML) models that are not fully understood and applies independently of the application domain.We offer a case study of reproducibility that highlights the challenges encountered when attempting to reproduce analyzes obtained with Machine Learning methods in materials informatics. Our study explores prediction results obtained with ML models and issues in training data serving as input. We discuss challenges related to theory-driven and numerical errors in training data, lack of reproducibility across platforms and versions, and effects of randomness when varying hyperparameters. In addition to model accuracy, a main metric of interest in the ML community, our results show that model sensitivity may be equally important for applying ML in domain applications such a materials science.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data

This study presents an AI-based constitutive modelling framework wherein the prediction model directly learns from triaxial testing data by combining discrete element modelling (DEM) and deep learning. A constitutive learning strategy is proposed based on the generally accepted frame-indifference assumption in constructing material constitutive models. The low-dimensional principal stress-strain sequence pairs, measured from discrete element modelling of triaxial testing, are used to train recurrent neural networks, and then the predicted principal stress sequence is augmented to other high-dimensional or general stress tensor via coordinate transformation. Through detailed hyperparameter investigations, it is found that long short-term memory (LSTM) and gated recurrent unit (GRU) networks have similar prediction performance in constitutive modelling problems, and both satisfactorily predict the stress responses of granular materials subjected to a given unseen strain path. Furthermore, the unique merits and ongoing challenges of data-driven constitutive models for granular materials are discussed.

42 ENGINEERING↗
DOE OSTI2022

Scalable deep learning for watershed model calibration

Watershed models such as the Soil and Water Assessment Tool (SWAT) consist of high-dimensional physical and empirical parameters. These parameters often need to be estimated/calibrated through inverse modeling to produce reliable predictions on hydrological fluxes and states. Existing parameter estimation methods can be time consuming, inefficient, and computationally expensive for high-dimensional problems. In this paper, we present an accurate and robust method to calibrate the SWAT model (i.e., 20 parameters) using scalable deep learning (DL). We developed inverse models based on convolutional neural networks (CNN) to assimilate observed streamflow data and estimate the SWAT model parameters. Scalable hyperparameter tuning is performed using high-performance computing resources to identify the top 50 optimal neural network architectures. We used ensemble SWAT simulations to train, validate, and test the CNN models. We estimated the parameters of the SWAT model using observed streamflow data and assessed the impact of measurement errors on SWAT model calibration. We tested and validated the proposed scalable DL methodology on the American River Watershed, located in the Pacific Northwest-based Yakima River basin. Our results show that the CNN-based calibration is better than two popular parameter estimation methods (i.e., the generalized likelihood uncertainty estimation [GLUE] and the dynamically dimensioned search [DDS], which is a global optimization algorithm). For the set of parameters that are sensitive to the observations, our proposed method yields narrower ranges than the GLUE method but broader ranges than values produced using the DDS method within the sampling range even under high relative observational errors. The SWAT model calibration performance using the CNNs, GLUE, and DDS methods are compared using R 2 and a set of efficiency metrics, including Nash-Sutcliffe, logarithmic Nash-Sutcliffe, Kling-Gupta, modified Kling-Gupta, and non-parametric Kling-Gupta scores, computed on the observed and simulated watershed responses. The best CNN-based calibrated set has scores of 0.71, 0.75, 0.85, 0.85, 0.86, and 0.91. The best DDS-based calibrated set has scores of 0.62, 0.69, 0.8, 0.77, 0.79, and 0.82. The best GLUE-based calibrated set has scores of 0.56, 0.58, 0.71, 0.7, 0.71, and 0.8. The scores above show that the CNN-based calibration leads to more accurate low and high streamflow predictions than the GLUE and DDS sets. Our research demonstrates that the proposed method has high potential to improve our current practice in calibrating large-scale integrated hydrologic models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Machine-learning based approach to examine ecological processes influencing the diversity of riverine dissolved organic matter composition

Dissolved organic matter (DOM) assemblages in freshwater rivers are formed from mixtures of simple to complex compounds that are highly variable across time and space. These mixtures largely form due to the environmental heterogeneity of river networks and the contribution of diverse allochthonous and autochthonous DOM sources. Most studies are, however, confined to local and regional scales, which precludes an understanding of how these mixtures arise at large, e.g., continental, spatial scales. The processes contributing to these mixtures are also difficult to study because of the complex interactions between various environmental factors and DOM. Here we propose the use of machine learning (ML) approaches to identify ecological processes contributing toward mixtures of DOM at a continental-scale. We related a dataset that characterized the molecular composition of DOM from river water and sediment with Fourier-transform ion cyclotron resonance mass spectrometry to explanatory physicochemical variables such as nutrient concentrations and stable water isotopes ( 2 H and 18 O). Using unsupervised ML, distinctive clusters for sediment and water samples were identified, with unique molecular compositions influenced by environmental factors like terrestrial input and microbial activity. Sediment clusters showed a higher proportion of protein-like and unclassified compounds than water clusters, while water clusters exhibited a more diversified chemical composition. We then applied a supervised ML approach, involving a two-stage use of SHapley Additive exPlanations (SHAP) values. In the first stage, SHAP values were obtained and used to identify key physicochemical variables. These parameters were employed to train models using both the default and subsequently tuned hyperparameters of the Histogram-based Gradient Boosting (HGB) algorithm. The supervised ML approach, using HGB and SHAP values, highlighted complex relationships between environmental factors and DOM diversity, in particular the existence of dams upstream, precipitation events, and other watershed characteristics were important in predicting higher chemical diversity in DOM. Our data-driven approach can now be used more generally to reveal the interplay between physical, chemical, and biological factors in determining the diversity of DOM in other ecosystems.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Robust Learning with Implicit Residual Networks

In this effort, we propose a new deep architecture utilizing residual blocks inspired by implicit discretization schemes. As opposed to the standard feed-forward networks, the outputs of the proposed implicit residual blocks are defined as the fixed points of the appropriately chosen nonlinear transformations. We show that this choice leads to the improved stability of both forward and backward propagations, has a favorable impact on the generalization power, and allows for control the robustness of the network with only a few hyperparameters. In addition, the proposed reformulation of ResNet does not introduce new parameters and can potentially lead to a reduction in the number of required layers due to improved forward stability. Finally, we derive the memory-efficient training algorithm, propose a stochastic regularization technique, and provide numerical results in support of our findings.

Reshniak, Viktor↗
DOE OSTI2024

Contrastive Machine Learning with Gamma Spectroscopy Data Augmentations for Detecting Shielded Radiological Material Transfers

Data analysis techniques can be powerful tools for rapidly analyzing data and extracting information that can be used in a latent space for categorizing observations between classes of data. Machine learning models that exploit learned data relationships can address a variety of nuclear nonproliferation challenges like the detection and tracking of shielded radiological material transfers. The high resource cost of manually labeling radiation spectra is a hindrance to the rapid analysis of data collected from persistent monitoring and to the adoption of supervised machine learning methods that require large volumes of curated training data. Instead, contrastive self-supervised learning on unlabeled spectra can enhance models that are built on limited labeled radiation datasets. This work demonstrates that contrastive machine learning is an effective technique for leveraging unlabeled data in detecting and characterizing nuclear material transfers demonstrated on radiation measurements collected at an Oak Ridge National Laboratory testbed, where sodium iodide detectors measure gamma radiation emitted by material transfers between the High Flux Isotope Reactor and the Radiochemical Engineering Development Center. Label-invariant data augmentations tailored for gamma radiation detection physics are used on unlabeled spectra to contrastively train an encoder, learning a complex, embedded state space with self-supervision. A linear classifier is then trained on a limited set of labeled data to distinguish transfer spectra between byproducts and tracked nuclear material using representations from the contrastively trained encoder. The optimized hyperparameter model achieves a balanced accuracy score of 80.30%. Any given model—that is, a trained encoder and classifier—shows preferential treatment for specific subclasses of transfer types. Regardless of the classifier complexity, a supervised classifier using contrastively trained representations achieves higher accuracy than using spectra when trained and tested on limited labeled data.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Factors Controlling a Synthetic Aperture Radar (SAR) Derived Root-Zone Soil Moisture Product over The Seward Peninsula of Alaska

Root-zone soil moisture exerts a fundamental control on vegetation, energy balance, and the carbon cycle in Arctic ecosystems, but it is still not well understood in vast, remote, and understudied regions of discontinuous permafrost. The root-zone soil moisture product (30 m resolution) used in this analysis was retrieved from a time-series P-Band (420–440 MHz) synthetic aperture radar (SAR) backscatter observations (August 2017 & October 2017). While similar approaches have been taken to retrieve surface (0 cm to 5 cm) soil moisture from L-Band (1.2 GHz) SAR backscatter, this is one of the first known attempts at reaching the root-zone in permafrost regions. Here, we analyze secondary factors (excluding primary factors, such as precipitation) controlling summer (August) soil moisture at depths of 6 cm, 12 cm, and 20 cm over a 4500 km2 area on the Seward Peninsula of Alaska. Using a random forest model, we quantify the impact of topography, vegetation, and meteorological factors on soil moisture distributions. In developing the random forest model, we explore a variety of feature scales (30 m, 60 m, 90 m, 120 m, 180 m, and 240 m), tune hyperparameters (the structure of individual decision trees making up the ensemble including the number and depth of trees), and perform the final feature selection using cross-validated recursive feature elimination. Results suggest that root-zone soil moisture on the Seward Peninsula is primarily controlled by vegetation at 6 cm, but deeper in the soil column topography and meteorological factors, such as predominant winter wind direction and summer insolation, play a larger role. The random forest model accounts for 40% to 60% of the variation observed (R2 = 0.44 at 6 cm, R2 = 0.52 at 12 cm, R2 = 0.58 at 20 cm). These results indicate that vegetation is the dominant control on soil moisture shallow in the soil column, but the impact of vegetation does not extend to deeper layers retrieved from P-Band SAR backscatter.

Dann, Julian↗
DOE OSTI2022

A Comparative Study of Machine Learning Algorithms for Industry-Specific Freight Generation Model

According to Bureau of Transportation Statistics, the U.S. transportation system handled 14,329 million ton-miles of freight per day in 2020. Understanding the generation of these freight shipments is crucial for transportation researchers, planners, and policymakers to design and plan for a more efficient and connected freight transportation system. Traditionally, the freight generation modeling has been based on Ordinary Least Square (OLS) regression, although more advanced Machine Learning (ML) algorithms have been evaluated and proven to have excellent performance in various transportation applications in recent years. Furthermore, one modeling approach applied for one industry might not always be applicable for another as their freight generation logics can be quite different. The objective of this study is to apply and evaluate alternative ML algorithms in the estimation of freight generation for each of 45 industry types. Seven alternative ML algorithms, along with the base OLS regression, were evaluated and compared. In addition, the study considered different combinations of variables in both the original and logarithmic form as well as hyperparameters of those ML algorithms in the model selection for each industry type. The results showed statistically significant improvements in the root mean square error reduction by the alternative ML algorithms over the OLS for over 80% of cases. The study suggests utilizing the alternative ML algorithms can reduce the root mean square error by about 30%, depending on industry types.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Recent Progress on Surface Water Quality Models Utilizing Machine Learning Techniques

Surface waterbodies are heavily exposed to pollutants caused by natural disasters and human activities. Empowering sensor technologies in water quality monitoring, sufficient measurements have become available to develop machine learning (ML) models. Numerous ML models have quickly been adopted to predict water quality indicators in various surface waterbodies. This paper reviews 78 recent articles from 2022 to October 2024, categorizing water quality models utilizing ML into three groups: Point-to-Point (P2P), which estimates the current target value based on other measurements at the same time point; Sequence-to-Point (S2P), which utilizes previous time series data to predict the target value at one time point ahead; and Sequence-to-Sequence (S2S), which uses previous time series data to forecast sequential target values in the future. The ML models used in each group are classified and compared according to water quality indicators, data availability, and model performance. Widely used strategies for improving performance, including feature engineering, hyperparameter tuning, and transfer learning, are recognized and described to enhance model effectiveness. The interpretability limitations of ML applications are discussed. This review provides a perspective on emerging ML for surface water quality models.

machine learning (ML)↗
DOE OSTI2023

Early Fault Detection in Particle Accelerator Power Electronics Using Ensemble Learning

Early fault detection and fault prognosis are crucial to ensure efficient and safe operations of complex engineering systems such as the Spallation Neutron Source (SNS) and its power electronics (high voltage converter modulators). Following an advanced experimental facility setup that mimics SNS operating conditions, the authors successfully conducted 21 early fault detection experiments, where fault precursors are introduced in the system to a degree enough to cause degradation in the waveform signals, but not enough to reach a real fault. Nine different machine learning techniques based on ensemble trees, convolutional neural networks, support vector machines, and hierarchical voting ensembles are proposed to detect the fault precursors. Although all 9 models have shown a perfect and identical performance during the training and testing phase, the performance of most models has decreased in the next test phase once they got exposed to realworld data from the 21 experiments. The hierarchical voting ensemble, which features multiple layers of diverse models, maintains a distinguished performance in early detection of the fault precursors with 95% success rate (20/21 tests), followed by adaboost and extremely randomized trees with 52% and 48% success rates, respectively. The support vector machine models were the worst with only 24% success rate (5/21 tests). The study concluded that a successful implementation of machine learning in the SNS or particle accelerator power systems would require a major upgrade in the controller and the data acquisition system to facilitate streaming and handling big data for the machine learning models. In addition, this study shows that the best performing models were diverse and based on the ensemble concept to reduce the bias and hyperparameter sensitivity of individual models.

43 PARTICLE ACCELERATORS↗
DOE OSTI2020

Dynamical Mass Estimates of the β Pictoris Planetary System through Gaussian Process Stellar Activity Modeling

Nearly 15 yr of radial velocity (RV) monitoring and direct imaging enabled the detection of two giant planets orbiting the young, nearby star β Pictoris. The δ Scuti pulsations of the star, which overwhelm planetary signals, need to be carefully suppressed. In this work, we independently revisit the analysis of the RV data following a different approach than available in the literature to model the activity of the star. We show that a Gaussian process (GP) with a stochastically driven damped harmonic oscillator kernel can model the δ Scuti pulsations. It provides similar results to parametric models but with a simpler framework, using only three hyperparameters. It also enables us to model poorly sampled RV data that were excluded from previous analyses, hence extending the RV baseline by nearly five years. Altogether, the orbit and mass of both planets can be constrained from RV only, which was not possible with the parametric modeling. To characterize the system more accurately, we also perform a joint fit of all available relative astrometry and RV data. Our orbital solutions for β Pic b favor a low eccentricity of 0.029$_{−0.024}^{+0.061}$ and a relatively short period of 21.1$_{−0.8}^{+2.0}$ yr. The orbit of β Pic c is eccentric with 0.206$_{−0.063}^{+0.074}$ with a period of 3.36 ± 0.03 yr. We find model-independent masses of 11.7 ± 1.4 and 8.5 ± 0.5 M Jup for β Pic b and c, respectively, assuming coplanarity. The mass of β Pic b is consistent with the hottest start evolutionary models, at an age of 25 ± 3 Myr. A direct detection of β Pic c would provide a second calibration measurement in a coeval system.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

ELT Imaging of MWC 297 from the 23 m LBTI: Complex Disk Structure and a Companion Candidate

Herbig Ae/Be stars represent the early outcomes of star formation and the initial stages of planet formation at intermediate stellar masses. Understanding both of these processes requires detailed characterization of their disk structures and companion frequencies. We present new 3.7 μm imaging of the Herbig Be star MWC 297 from nonredundant masking observations on the phase-controlled, 23 m Large Binocular Telescope Interferometer. The images reveal complex disk structure on the scales of several au, as well as a companion candidate. We discuss physical interpretations for these features and demonstrate that the imaging results are independent of choices such as priors, regularization hyperparameters, and error-bar estimates. With an angular resolution of ∼17 mas, these data provide the first robust Extremely Large Telescope–resolution view of a distant young star.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

Star–Galaxy Image Separation with Computationally Efficient Gaussian Process Classification

Abstract We introduce a novel method for discerning optical telescope images of stars from those of galaxies using Gaussian processes (GPs). Although applications of GPs often struggle in high-dimensional data modalities such as optical image classification, we show that a low-dimensional embedding of images into a metric space defined by the principal components of the data suffices to produce high-quality predictions from real large-scale survey data. We develop a novel method of GP classification hyperparameter training that scales approximately linearly in the number of image observations, which allows for application of GP models to large-size Hyper Suprime-Cam Subaru Strategic Program data. In our experiments, we evaluate the performance of a principal component analysis embedded GP predictive model against other machine-learning algorithms, including a convolutional neural network and an image photometric morphology discriminator. Our analysis shows that our methods compare favorably with current methods in optical image classification while producing posterior distributions from the GP regression that can be used to quantify object classification uncertainty. We further describe how classification uncertainty can be used to efficiently parse large-scale survey imaging data to produce high-confidence object catalogs.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Possibilities and Limitations of Kinematically Identifying Stars from Accreted Ultra-faint Dwarf Galaxies

Abstract The Milky Way has accreted many ultra-faint dwarf galaxies (UFDs), and stars from these galaxies can be found throughout our Galaxy today. Studying these stars provides insight into galaxy formation and early chemical enrichment, but identifying them is difficult. Clustering stellar dynamics in 4D phase space ( E , L z , J r , J z ) is one method of identifying accreted structure that is currently being utilized in the search for accreted UFDs. We produce 32 simulated stellar halos using particle tagging with the Caterpillar simulation suite and thoroughly test the abilities of different clustering algorithms to recover tidally disrupted UFD remnants. We perform over 10,000 clustering runs, testing seven clustering algorithms, roughly twenty hyperparameter choices per algorithm, and six different types of data sets each with up to 32 simulated samples. Of the seven algorithms, HDBSCAN most consistently balances UFD recovery rates and cluster realness rates. We find that, even in highly idealized cases, the vast majority of clusters found by clustering algorithms do not correspond to real accreted UFD remnants and we can generally only recover 6% of UFDs remnants at best. These results focus exclusively on groups of stars from UFDs, which have weak dynamic signatures compared to the background of other stars. The recoverable UFD remnants are those that accreted recently, z accretion ≲ 0.5. Based on these results, we make recommendations to help guide the search for dynamically linked clusters of UFD stars in observational data. We find that real clusters generally have higher median energy and J r , providing a way to help identify real versus fake clusters. We also recommend incorporating chemical tagging as a way to improve clustering results.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Hierarchical Inference of the Lensing Convergence from Photometric Catalogs with Bayesian Graph Neural Networks

Abstract We present a Bayesian graph neural network (BGNN) that can estimate the weak lensing convergence ( κ ) from photometric measurements of galaxies along a given line of sight (LOS). The method is of particular interest in strong gravitational time-delay cosmography (TDC), where characterizing the “external convergence” ( κ ext ) from the lens environment and LOS is necessary for precise Hubble constant ( H 0 ) inference. Starting from a large-scale simulation with a κ resolution of ∼1′, we introduce fluctuations on galaxy–galaxy lensing scales of ∼1″ and extract random sight lines to train our BGNN. We then evaluate the model on test sets with varying degrees of overlap with the training distribution. For each test set of 1000 sight lines, the BGNN infers the individual κ posteriors, which we combine in a hierarchical Bayesian model to yield constraints on the hyperparameters governing the population. For a test field well sampled by the training set, the BGNN recovers the population mean of κ precisely and without bias (within the 2 σ credible interval), resulting in a contribution to the H 0 error budget well under 1%. In the tails of the training set with sparse samples, the BGNN, which can ingest all available information about each sight line, extracts a stronger κ signal compared to a simplified version of the traditional method based on matching galaxy number counts, which is limited by sample variance. Our hierarchical inference pipeline using BGNNs promises to improve the κ ext characterization for precision TDC. The code is available as a public Python package, Node to Joy ⏬ .

79 ASTRONOMY AND ASTROPHYSICS↗
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 OSTI2020

Extended Physics-Informed Neural Networks (XPINNs): A Generalized Space-Time Domain Decomposition Based Deep Learning Framework for Nonlinear Partial Differential Equations

Here we propose a generalized space-time domain decomposition approach for the physics-informed neural networks (PINNs) to solve nonlinear partial differential equations (PDEs) on arbitrary complex-geometry domains. The proposed framework, named eXtended PINNs ( X P I N N s ), further pushes the boundaries of both PINNs as well as conservative PINNs (cPINNs), which is a recently proposed domain decomposition approach in the PINN framework tailored to conservation laws. Compared to PINN, the XPINN method has large representation and parallelization capacity due to the inherent property of deployment of multiple neural networks in the smaller subdomains. Unlike cPINN, XPINN can be extended to any type of PDEs. Moreover, the domain can be decomposed in any arbitrary way (in space and time), which is not possible in cPINN. Thus, XPINN offers both space and time parallelization, thereby reducing the training cost more effectively. In each subdomain, a separate neural network is employed with optimally selected hyperparameters, e.g., depth/width of the network, number and location of residual points, activation function, optimization method, etc. A deep network can be employed in a subdomain with complex solution, whereas a shallow neural network can be used in a subdomain with relatively simple and smooth solutions. We demonstrate the versatility of XPINN by solving both forward and inverse PDE problems, ranging from one-dimensional to three-dimensional problems, from time-dependent to time-independent problems, and from continuous to discontinuous problems, which clearly shows that the XPINN method is promising in many practical problems. The proposed XPINN method is the generalization of PINN and cPINN methods, both in terms of applicability as well as domain decomposition approach, which efficiently lends itself to parallelized computation. The XPINN code is available on h t t p s : / / g i t h u b . c o m / A m e y a J a g t a p / X P I N N s .

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Technical note: Deep learning for creating surrogate models of precipitation in Earth system models

Abstract. We investigate techniques for using deep neural networks to produce surrogatemodels for short-term climate forecasts. A convolutional neural network istrained on 97 years of monthly precipitation output from the 1pctCO2 run (theCO 2 concentration increases by 1 % per year) simulated by the second-generation Canadian Earth System Model (CanESM2). The neural network clearly outperforms a persistence forecast anddoes not show substantially degraded performance even when the forecast lengthis extended to 120 months. The model is prone to underpredicting precipitationin areas characterized by intense precipitation events. Scheduled sampling(forcing the model to gradually use its own past predictions rather than groundtruth) is essential for avoiding amplification of early forecasting errors.However, the use of scheduled sampling also necessitates preforecasting(generating forecasts prior to the first forecast date) to obtain adequateperformance for the first few prediction time steps. We document the trainingprocedures and hyperparameter optimization process for researchers who wish toextend the use of neural networks in developing surrogate models.

54 ENVIRONMENTAL SCIENCES↗