Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed training”

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 163 records · Page 9

MO-FG-303-03: Demonstration of Universal Knowledge-Based 3D Dose Prediction

Purpose: To demonstrate a knowledge-based 3D dose prediction methodology that can accurately predict achievable radiotherapy distributions. Methods: Using previously treated plans as input, an artificial neural network (ANN) was trained to predict 3D dose distributions based on 14 patient-specific anatomical parameters including the distance (r) to planning target volume (PTV) boundary, organ-at-risk (OAR) boundary distances, and angular position ( θ,φ). 23 prostate and 49 stereotactic radiosurgery (SRS) cases with ≥1 nearby OARs were studied. All were planned with volumetric-modulated arc therapy (VMAT) to prescription doses of 81Gy for prostate and 12–30Gy for SRS. Site-specific ANNs were trained using all prostate 23 plans and using a 24 randomly-selected subset for the SRS model. The remaining 25 SRS plans were used to validate the model. To quantify predictive accuracy, the dose difference between the clinical plan and prediction were calculated on a voxel-by-voxel basis δD(r,θ,φ)=Dclin(r,θ,φ)-Dpred(r, θ,φ). Grouping voxels by boundary distance, the mean <δ Dr>=(1/N)Σ -θ,φ D(r,θ,φ) and inter-quartile range (IQR) quantified the accuracy of this method for deriving DVH estimations. The standard deviation (σ) of δ D quantified the 3D dose prediction error on a voxel-by-voxel basis. Results: The ANNs were highly accurate in predictive ability for both prostate and SRS plans. For prostate, <δDr> ranged from −0.8% to +0.6% (max IQR=3.8%) over r=0–32mm, while 3D dose prediction accuracy averaged from σ=5–8% across the same range. For SRS, from r=0–34mm the training set <δDr> ranged from −3.7% to +1.5% (max IQR=4.4%) while the validation set <δDr> ranged from −2.2% to +5.8% (max IQR=5.3%). 3D dose prediction accuracy averaged σ=2.5% for the training set and σ=4.0% over the same interval. Conclusion: The study demonstrates this technique’s ability to predict achievable 3D dose distributions for VMAT SRS and prostate. Future investigations will attempt to separate and quantify modeling error and plan quality variations in the statistical noise. Research funding support from Varian Medical Systems.

60 APPLIED LIFE SCIENCES↗

Understanding and Predicting the Spatially Resolved Adsorption Properties of Nanoporous Materials

Using knowledge from statistical thermodynamics and crystallography, we develop an image–image translation model, called SorbIIT, that uses three-dimensional grids of adsorbate–adsorbent interaction energies as input to predict the spatially resolved loading surface of nanoporous materials over a broad range of temperatures and pressures. SorbIIT consists of a closed-form differential model for loading-surface prediction and a U-Net to generate spatial differential distributions from the energy grids. SorbIIT is trained using the energy grids and adsorbate distributions (obtained from high-throughput simulations) of 50 synthesized and 70 hypothetical zeolites and applied for predicting the adsorption of carbon dioxide, hydrogen sulfide, n-butane, 2-methylpropane, krypton, and xenon in other zeolites from 256 to 400 K. In conclusion, employing a quadratic isotherm model for the local differentiation, SorbIIT yields mean R 2 values of 0.998 for total adsorption and 0.6904 for local adsorption with a resolution of 0.2 Å, and a value of 0.721 for the structural similarity of the local loading distribution.

Sun, Yangzesheng↗

Fast and efficient identification of anomalous galaxy spectra with neural density estimation

ABSTRACT Current large-scale astrophysical experiments produce unprecedented amounts of rich and diverse data. This creates a growing need for fast and flexible automated data inspection methods. Deep learning algorithms can capture and pick up subtle variations in rich data sets and are fast to apply once trained. Here, we study the applicability of an unsupervised and probabilistic deep learning framework, the probabilistic auto-encoder, to the detection of peculiar objects in galaxy spectra from the SDSS survey. Different to supervised algorithms, this algorithm is not trained to detect a specific feature or type of anomaly, instead it learns the complex and diverse distribution of galaxy spectra from training data and identifies outliers with respect to the learned distribution. We find that the algorithm assigns consistently lower probabilities (higher anomaly score) to spectra that exhibit unusual features. For example, the majority of outliers among quiescent galaxies are E+A galaxies, whose spectra combine features from old and young stellar population. Other identified outliers include LINERs, supernovae, and overlapping objects. Conditional modelling further allows us to incorporate additional information. Namely, we evaluate the probability of an object being anomalous given a certain spectral class, but other information such as metrics of data quality or estimated redshift could be incorporated as well. We make our code publicly available.

Böhm, Vanessa↗

Evaluating Air-Source Heat Pumps with Distributed Energy Resources in REopt

This training introduces attendees to NREL's REopt(R) web tool and its capabilities in evaluating air-source heat pumps (ASHP) alongside distributed energy resources (DERs). Designed for energy managers, facility operators, and sustainability managers, this session will explore how REopt provides techno-economic analysis to assess the financial viability of ASHP while identifying synergies with solar PV and battery storage. Participants will gain practical insights into using REopt to optimize energy strategies, reduce costs, and improve site resilience.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Error-mitigated data-driven circuit learning on noisy quantum hardware

Application-level benchmarks measure how well a quantum device performs meaningful calculations. In the case of parameterized circuit training, the computational task is the preparation of a target quantum state via optimization over a loss landscape. This is complicated by various sources of noise, fixed hardware connectivity, and generative modeling, the choice of target distribution. Gradient-based training has become a useful benchmarking task for noisy intermediate-scale quantum computers because of the additional requirement that the optimization step uses the quantum device to estimate the loss function gradient. In this work, we use gradient-based data-driven circuit learning to qualitatively evaluate the performance of several superconducting platform devices and present results that show how error mitigation can improve the training of quantum circuit Born machines with 28 tunable parameters.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Once again on the duration of nuclear gamma-ray-emission and gamma-ray-absorption processes

In addition to previous indications of a long-term character of the processes of gamma-ray emission and absorption by nuclei, another two simple arguments in support of this picture are presented. It is shown that the Fourier integral for a short wave train is a frequency distribution whose width is many orders of magnitude larger than actual intrinsic widths of gamma lines. The uncertainty in the photon spatial position is found to be about {tau}c, which is the length of the wave train emitted by a nucleus over the average lifetime {tau} of this nucleus in an excited state.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Geometric Measures of Trustworthiness for Machine Learning Predictions

his report details the findings from the research and investigation of Geometric Measures of Trustworthiness for Machine Learning Predictions. We explored the trustworthiness of machine learning (ML) models’ predictions using geometric measures to quantify the similarity of a query point with the training data. Predictive uncertainty in ML can originate from at least three sources: (1) Model uncertainty, which represents the uncertainty in model form (e.g. decision tree, vs neural network) and estimating the model parameters from the training data, (2) Data uncertainty, which represents the natural complexities of the data such as class overlap and inherent noise, and (3) Distributional uncertainty, which represents the mismatch between the training and operational distributions. The proposed measures focus on measuring and explaining the data and distributional uncertainties by measuring the relationships of operational data with the training data.

97 MATHEMATICS AND COMPUTING↗

Energy Efficient Stochastic-Based Deep Spiking Neural Networks for Sparse Datasets

With large deep neural networks (DNNs) neces- sary to solve complex and data-intensive problems, energy efficiency is a key bottleneck for effectively deploying DL in the real world. Deep spiking NNs have gained much research at- tention recently due to the interest in building biological neural networks and the availability of neuromorphic platforms, which can be orders of magnitude more energy efficient compared to CPUs and GPUs. Although spiking NNs have proven to be an efficient technique for solving many machine learning and computer vision problems, to the best of our knowledge, this is the first attempt to adapt spiking NNs to sparse datasets. In this paper, we study the behaviour of spiking NNs in handling NLP datasets and the sparsity in their data representation. Then, we propose a novel framework for spiking NN using the concept of stochastic computing. Specifically, instead of generating spike trains with firing rates proportional to the intensity of each value in the feature set separately, the whole feature set is treated as a distribution function and a stochastic spiking train that follow this distribution is generated. This framework reduces the connectivity between NN layers from O(N) to O(logN). Also, it encodes input data differently and make suitable to handle sparse datasets. Finally, the framework achieves high energy efficiency since it uses Integrate and Fire neurons same as conventional spiking NNs. The results show that our proposed stochastic-based SNN achieves nearly the same accuracy as the original DNN on MNIST dataset, and it has better performance than state- of-the-art SNN. Besides that stochastic-based SNN is energy efficient, where the fully connected DNN, the conventional SNN, and the data normalized SNN consume 38.24, 1.83, and 1.85 -times more energy than the stochastic-based SNN, respectively. For sparse datasets, including IMDb and In-House clinical datasets, stochastic-based SNN achieves performance comparable to that of the conventional DNN. However, the conventional spiking NN has a significant decline in classification accuracy.

Alawad, Mohammed↗

HydraGNN_Predictive_GFM_2024 - Ensemble of predictive graph foundation models for ground state atomistic materials modeling

We provide the ensemble of fifteen pre-trained graph foundation models (GFMs) for atomistic materials modeling applications. Each one of the fifteen GFMs has been trained on five open-source datasets that (once aggregated) amount to over 154 million atomistic structures, which cover over two-thirds of the natural elements of the periodic table and that comprises a broad set of organic and inorganic compounds. This vast set of atomistic structures comprises ground state configurations that are dynamically stable (i.e., equilibrated structures with atomic forces approximately close to zero values) as well as dynamically unstable structures (i.e., non-equilibrium structures with non-negligible non-zero values of atomic forces). The ensemble of datasets aggregated does NOT include excited states. The datasets have been curated to remove atomistic structures with spectral norm of the force tensor above 100 eV/angstrom. Moreover, a linear term of the energy was computed for each dataset using a linear regression model that uses the chemical concentration of each natural element as regressor. The linear term predicted by the linear regression model has been subtracted from each original energy value to perform a re-alignment of the energy values across different electronic structures approximation theories performed to generate the diverse multi-source, multi-fidelity datasets. The folder "ADIOS_files" 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 "ADIOS_files" directory contains 6 sub-directories named as follows: - ANI1x-v3.bp - MPTrj-v3.bp - OC2020-20M-v3.bp - OC2020-v3.bp - OC2022-v3.bp - qm7x-v3.bp Each sub-directory contains the pre-processed datasets converted in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used to the development, training, and performance testing of the ensemble go predictive graph foundation models. Each GFM was developed using HydraGNN (https://github.com/ORNL/HydraGNN) as underlying graph neural network (GNN) architecture. The multi-task learning (MTL) capability of HydraGNN was used to simultaneously train the GFMs on labeled values for direct predictions of energy (a total system property of an atomistic structure that measures the chemical stability) and atomic forces (an atomic level property of an atomistic structure that measures the dynamical stability). The hyper parameters of the GFM have been tuned using scalable hyperparameter optimization (HPO) algorithms implemented in the software DeepHyper (https://github.com/deephyper/deephyper). The pre-training of each HPO trial was performed using distributed data parallelism (DDP) to scale the training across 128 compute nodes of the exascale OLCF supercomputer Frontier. Each HPO trial was trained only for 10 epochs and an early stopping was performed to avoid wasting significant computational resources on GNN architectures that were clearly underperforming. For each HPO trial, the 'omnistat' tool developed by (AMD Research - Advanced Micro Device) was used to measure the total energy consumption in kWh. The ensemble of GFMs was obtained by selecting the fifteen best performing HPO trials. Four models have been selected for their clear advantage in accuracy, and these are the GFMs with IDs 229, 156, 147, 260. Additional eleven models have been selected based on judicious balance between accuracy and energy consumption needed for training, and these are the GFMs with IDs 165, 78, 137, 1, 175, 171, 181, 67, 179, 167, 351. Each selected GFM of the ensemble was continued to cumulate a total of at most 30 epochs. In some cases, the total number of epochs actually performed was les than 30 due to two combined factors: (1) the size of the GFM (i.e., the number of model parameters to train) and (2) the total wall-clock time for which the computational resources could be allocated on OLCF-Frontier. The "Ensemble_of_models" directory contains 15 sub-directories named as follows: - gfm_0.229 - gfm_0.156 - gfm_0.147 - gfm_0.260 - gfm_0.165 - gfm_0.78 - gfm_0.137 - gfm_0.1 - gfm_0.175 - gfm_0.171 - gfm_0.181 - gfm_0.67 - gfm_0.179 - gfm_0.167 - gfm_0.351 Each one of these sub-directories refers to one of the fifteen HPO trials that have been selected to continue the pre-training with at most 30 epochs. With each sub-directory associated with a specific HPO trial, the following files can be found: - config.json: file for argument parsing to develop and train an HydraGNN architecture - gfm_0.ID_epoch_N.pk: file with model parameters for HPO ID trial after N epochs of training The ensemble of fifteen GFM architectures was used for (1) ensemble averaging to stabilize the predictions of energy and atomic forces after pre-training for post-processing analysis and (2) ensemble uncertainty quantification (UQ). The code used to develop, pre-train, and load the pre-trained models for post-processing analysis is available on the ORNL-GitHub at the following link: https://github.com/ORNL/HydraGNN/tree/Predictive_GFM_2024

36 MATERIALS SCIENCE↗

Predicting images for the dynamics of stellar clusters ( π-DOC ): a deep learning framework to predict mass, distance, and age of globular clusters

ABSTRACT Dynamical mass estimates of simple systems such as globular clusters (GCs) still suffer from up to a factor of 2 uncertainty. This is primarily due to the oversimplifications of standard dynamical models that often neglect the effects of the long-term evolution of GCs. Here, we introduce a new approach to measure the dynamical properties of GCs, based on the combination of a deep-learning framework and the large amount of data from direct N-body simulations. Our algorithm, π-DOC (Predicting Images for the Dynamics Of stellar Clusters) is composed of two convolutional networks, trained to learn the non-trivial transformation between an observed GC luminosity map and its associated mass distribution, age, and distance. The training set is made of V-band luminosity and mass maps constructed as mock observations from N-body simulations. The tests on π-DOC demonstrate that we can predict the mass distribution with a mean error per pixel of 27 per cent, and the age and distance with an accuracy of 1.5 Gyr and 6 kpc, respectively. In turn, we recover the shape of the mass-to-light profile and its global value with a mean error of 12 per cent, which implies that we efficiently trace mass segregation. A preliminary comparison with observations indicates that our algorithm is able to predict the dynamical properties of GCs within the limits of the training set. These encouraging results demonstrate that our deep-learning framework and its forward modelling approach can offer a rapid and adaptable tool competitive with standard dynamical models.

Chardin, Jonathan↗

Fast and self-learning indoor airflow simulation based on in situ adaptive tabulation

Fast simulation for stratified indoor airflow distributions is desired for various applications, such as design of advanced indoor environments, emergency management, and coupled annual energy simulation for buildings with stratified air distributions. Reduced order models trained by pre-computed computational fluid dynamics results are fast, but their prediction may be inaccurate when applied for conditions outside the training domain. To overcome this limitation, we propose in this paper a fast and self-learning model based on an in situ adaptive tabulation (ISAT) algorithm, which is trained by a fast fluid dynamics (FFD) model as an example. The idea is that the ISAT will retrieve the solutions from an existing data set if the estimated prediction error is within a pre-defined tolerance. Otherwise, the ISAT will execute the FFD simulation, which is accelerated by running in parallel on a graphics processing unit, for a full-scale simulation. This paper systematically investigates the feasibility of the ISAT for indoor airflow simulations by presenting the ISAT-FFD implementation alongside results related to its overall performance. Using a stratified indoor airflow as an example, we evaluated how the training time of ISAT was impacted by four factors (training methods, error tolerances, number of inputs, and number of outputs). Then we demonstrated that a trained ISAT model can predict the key information for inputs both inside and outside the training domain. The ISAT was able to answer query points both inside and close to training domain using retrieve actions within a time less than 0.001 s for each query. In conclusion, we provided suggestions for using the ISAT for building applications.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Optimizing aircraft flows at airports using data driven predicted capabilities

A method for safe and efficient use of airport runway capacity includes receiving, at an air traffic control system at an airport, airport data related to movement areas of the airport, time data related to a time period, aircraft data related to a plurality of aircraft expected to operate into and out of the airport during the time period, and environmental data related to environmental conditions predicted for the airport during the time period. The method further includes computing a probability distribution for inter-aircraft spacing by applying the airport data, the time data, the aircraft data, and the environmental data to a trained Bayesian network, producing the probability distribution for the inter-aircraft spacing as an output observation of the trained Bayesian network, and, using the probability distribution and a confidence value, identifying an inter-aircraft spacing value for the plurality of aircraft expected to operate into and out of the airport during the time period.

Sweet, Douglas↗

SIDDA: SInkhorn Dynamic Domain Adaptation

Modern neural networks (NNs) often do not generalize well in the presence of a "covariate shift"; that is, in situations where the training and test data distributions differ, but the conditional distribution of classification labels remains unchanged. In such cases, NN generalization can be reduced to a problem of learning more domain-invariant features. Domain adaptation (DA) methods include a range of techniques aimed at achieving this; however, these methods have struggled with the need for extensive hyperparameter tuning, which then incurs significant computational costs. In this work, we introduce SIDDA, an out-of-the-box DA training algorithm built upon the Sinkhorn divergence, that can achieve effective domain alignment with minimal hyperparameter tuning and computational overhead. We demonstrate the efficacy of our method on multiple simulated and real datasets of varying complexity, including simple shapes, handwritten digits, and real astronomical observations. SIDDA is compatible with a variety of NN architectures, and it works particularly well in improving classification accuracy and model calibration when paired with equivariant neural networks (ENNs). We find that SIDDA enhances the generalization capabilities of NNs, achieving up to a ≈40% improvement in classification accuracy on unlabeled target data. We also study the efficacy of DA on ENNs with respect to the varying group orders of the dihedral group DN, and find that the model performance improves as the degree of equivariance increases. Finally, we find that SIDDA enhances model calibration on both source and target data--achieving over an order of magnitude improvement in the ECE and Brier score. SIDDA's versatility, combined with its automated approach to domain alignment, has the potential to advance multi-dataset studies by enabling the development of highly generalizable models.

Pandya, Sneh [Northeastern U.]↗

Measuring the thermal and ionization state of the low- z IGM using likelihood free inference

ABSTRACT We present a new approach to measure the power-law temperature density relationship $T=T_0 (\rho/ \bar{\rho })^{\gamma -1}$ and the UV background photoionization rate $\Gamma _{{{{\rm H\, {\small I}}}}{}}$ of the intergalactic medium (IGM) based on the Voigt profile decomposition of the Ly α forest into a set of discrete absorption lines with Doppler parameter b and the neutral hydrogen column density $N_{\rm H\, {\small I}}$. Previous work demonstrated that the shape of the $b-N_{{{{\rm H\, {\small I}}}}{}}$ distribution is sensitive to the IGM thermal parameters T0 and γ, whereas our new inference algorithm also takes into account the normalization of the distribution, i.e. the line-density dN/dz, and we demonstrate that precise constraints can also be obtained on $\Gamma _{{{{\rm H\, {\small I}}}}{}}$. We use density-estimation likelihood-free inference (DELFI) to emulate the dependence of the $b-N_{{{{\rm H\, {\small I}}}}{}}$ distribution on IGM parameters trained on an ensemble of 624 nyx hydrodynamical simulations at z = 0.1, which we combine with a Gaussian process emulator of the normalization. To demonstrate the efficacy of this approach, we generate hundreds of realizations of realistic mock HST/COS data sets, each comprising 34 quasar sightlines, and forward model the noise and resolution to match the real data. We use this large ensemble of mocks to extensively test our inference and empirically demonstrate that our posterior distributions are robust. Our analysis shows that by applying our new approach to existing Ly α forest spectra at z ≃ 0.1, one can measure the thermal and ionization state of the IGM with very high precision ($\sigma _{\log T_0} \sim 0.08$ dex, σγ ∼ 0.06, and $\sigma _{\log \Gamma _{{{{\rm H\, {\small I}}}}{}}} \sim 0.07$ dex).

79 ASTRONOMY AND ASTROPHYSICS↗

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

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

79 ASTRONOMY AND ASTROPHYSICS↗

RLC4CLR (Reinforcement Learning Controller for Critical Load Restoration Problems)

RLC4CLR demonstrates using a reinforcement learning controller (RLC) to solve a critical load restoration (CLR) problem, which improves the grid resilience after a substation outage event. RLC4CLR consists of two parts. (1) RL environment: This environment encapsulates the CLR problem to be solved and provides interfacing functions to follow the standard OpenAI Gym format. A power system simulator, i.e., OpenDSS, is included to provide the power flow solution. Controller inputs and outputs (RL state and action) as well as the reward are defined in this environment as well. In summary, the RL environment is the problem formulation from which the RL agent can learn. (2) RL training script: The training script enables the RL agent to learn its control policy by interacting with the RL environment. For RL training, an open-sourced RL library, i.e., RLlib, is leveraged which is based on a distributed computing framework (Ray). The training script is designed to be able to be run on both local machine or the NREL HPC system. Other components of RLC4CLR include input data, e.g., grid model (standard IEEE test feeders), and other files used for results analysis.

Zhang, Xiangyu↗

Scale-up Unlearnable Examples Learning with High-performance Computing

Recent advancements in AI models, like ChatGPT, are structured to retain user interactions, which could inadvertently include sensitive healthcare data. In the healthcare field, particularly when radiologists use AI-driven diagnostic tools hosted on online platforms, there is a risk that medical imaging data may be repurposed for future AI training without explicit consent, spotlighting critical privacy and intellectual property concerns around healthcare data usage. Addressing these privacy challenges, a novel approach known as Unlearnable Examples (UEs) has been introduced, aiming to make data unlearnable to deep learning models. A prominent method within this area, called Unlearnable Clustering (UC), has shown improved UE performance with larger batch sizes but was previously limited by computational resources (e.g., a single workstation). To push the boundaries of UE performance with theoretically unlimited resources, we scaled up UC learning across various datasets using Distributed Data Parallel (DDP) training on the Summit supercomputer. Our goal was to examine UE efficacy at high-performance computing (HPC) levels to prevent unauthorized learning and enhance data security, particularly exploring the impact of batch size on UE’s unlearnability. Utilizing the robust computational capabilities of the Summit, extensive experiments were conducted on diverse datasets such as Pets, MedMNist, Flowers, and Flowers102. Our findings reveal that both overly large and overly small batch sizes can lead to performance instability and affect accuracy. However, the relationship between batch size and unlearnability varied across datasets, highlighting the necessity for tailored batch size strategies to achieve optimal data protection. The use of Summit’s high-performance GPUs, along with the efficiency of the DDP framework, facilitated rapid updates of model parameters and consistent training across nodes. Our results underscore the critical role of selecting appropriate batch sizes based on the specific characteristics of each dataset to prevent learning and ensure data security in deep learning applications. The source code is publicly available at https: // github. com/ hrlblab/ UE_ HPC .

Zhu, Yanfan [Vanderbilt University, Nashville, TN,↗

Experimental validation of a high fidelity Monte Carlo neutron transport model of the MIT graphite exponential pile

High-fidelity modeling and simulation were performed for the MIT graphite exponential pile (MGEP) using Monte Carlo neutron transport codes OpenMC and MCNP, and the results were validated by experimental data. The MGEP is being used as the test bed for the design of an autonomous control system for the pile's neutron flux distribution. The main contribution of this work is to generate the training data sets of neutron flux distributions with different locations of control rods that perturb the neutron flux profiles. First, code -to-code cross verification between OpenMC and MCNP was performed to ensure consistency of the numerical modeling within statistical uncertainties. To validate the accuracy of this high-fidelity model, a series of neutron flux measurements were conducted using a Helium-3 (He-3) neutron detector on a mobile platform that is placed inside the pile. Second, the neutron flux profiles were measured in four vertical layers of interest, and compared to the corresponding simulation results. The comparison results shows that the root mean square error is less than 2.5% in the two upper layers, and less than 4.5% in all four measured layers. Here the results validated the accuracy of the modeling and simulation. Finally, the relative change of the neutron flux profiles from moving control rods was analyzed, which identified the layer that has the best sensitivity regarding the control rods movements. Thus, this work identified and provided training data sets of both simulated and experimental neutron flux profiles in the most sensitive layer, paving the path forward to the real-time experimental demonstration of the autonomous control system.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗