Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “train”

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 559 records · Page 31
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Detecting Large Explosions With Machine Learning Models Trained on Synthetic Infrasound Data

Explosions produce low-frequency acoustic (infrasound) waves capable of propagating globally, but the spatio-temporal variability of the atmosphere makes detecting events difficult. Machine learning (ML) is well-suited to identify the subtle and nonlinear patterns in explosion infrasound signals, but a previous lack of ground-truth data inhibited training of generalized models. We introduce a physics-based method that propagates infrasound sources through realistic atmospheres to create 28,000 synthetic events, which are used to train ML classifiers. A simple artificial neural network and modern temporal convolutional network discriminate synthetic events from background noise with >90% accuracy and, more importantly, successfully identify the majority of real-world explosion signals recorded during the Humming Road Runner experiment. ML models trained entirely on physics-based synthetics advance explosion detection capabilities and make ML more viable to related fields lacking training data.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2024

A Novel Method to Train Classification Models for Structure Detection in In Situ Spacecraft Data

We present a method for creating spacecraft-like data which can be used to train Machine Learning (ML) models to detect and classify structures in in situ spacecraft data. First, we use the Grad-Shafranov equation to numerically solve for several magnetohydrostatic equilibria which are variations on a known analytic equilibrium. These equilibria are then used as the initial conditions for Particle-In-Cell simulations in which the structures of interest are observed and labeled. We then take one-dimensional slices through the simulations to replicate what a spacecraft collecting data from the simulation would observe. This sliced data then can be used as training data for the initial training of ML models intended for use on spacecraft data. We demonstrate the method applied to the problem of detecting small-scale plasmoids in the magnetotail, which is important for understanding complex magnetotail reconnection dynamics. The simple 1D classifier we train is able to detect more than 70% of the plasmoid points in the data set but also produces a large number of false positives. Our further work on this example problem is detailed, and further potential uses of the method are discussed.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Temporal dynamics of the multi-omic response to endurance exercise training

Regular exercise promotes whole-body health and prevents disease, but the underlying molecular mechanisms are incompletely understood. Here, the Molecular Transducers of Physical Activity Consortium profiled the temporal transcriptome, proteome, metabolome, lipidome, phosphoproteome, acetylproteome, ubiquitylproteome, epigenome and immunome in whole blood, plasma and 18 solid tissues in male and female Rattus norvegicus over eight weeks of endurance exercise training. The resulting data compendium encompasses 9,466 assays across 19 tissues, 25 molecular platforms and 4 training time points. Thousands of shared and tissue-specific molecular alterations were identified, with sex differences found in multiple tissues. Temporal multi-omic and multi-tissue analyses revealed expansive biological insights into the adaptive responses to endurance training, including widespread regulation of immune, metabolic, stress response and mitochondrial pathways. Many changes were relevant to human health, including non-alcoholic fatty liver disease, inflammatory bowel disease, cardiovascular health and tissue injury and recovery. The data and analyses presented in this study will serve as valuable resources for understanding and exploring the multi-tissue molecular effects of endurance training and are provided in a public repository (https://motrpac-data.org/).

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Adaptive hyperparameter updating for training restricted Boltzmann machines on quantum annealers

Restricted Boltzmann Machines (RBMs) have been proposed for developing neural networks for a variety of unsupervised machine learning applications such as image recognition, drug discovery, and materials design. The Boltzmann probability distribution is used as a model to identify network parameters by optimizing the likelihood of predicting an output given hidden states trained on available data. Training such networks often requires sampling over a large probability space that must be approximated during gradient based optimization. Quantum annealing has been proposed as a means to search this space more efficiently which has been experimentally investigated on D-Wave hardware. D-Wave implementation requires selection of an effective inverse temperature or hyperparameter (β) within the Boltzmann distribution which can strongly influence optimization. Here, we show how this parameter can be estimated as a hyperparameter applied to D-Wave hardware during neural network training by maximizing the likelihood or minimizing the Shannon entropy. We find both methods improve training RBMs based upon D-Wave hardware experimental validation on an image recognition problem. Neural network image reconstruction errors are evaluated using Bayesian uncertainty analysis which illustrate more than an order magnitude lower image reconstruction error using the maximum likelihood over manually optimizing the hyperparameter. The maximum likelihood method is also shown to out-perform minimizing the Shannon entropy for image reconstruction.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Teacher-student training improves the accuracy and efficiency of machine learning interatomic potentials

Machine learning interatomic potentials (MLIPs) are revolutionizing the field of molecular dynamics (MD) simulations. Recent MLIPs have tended towards more complex architectures trained on larger datasets. The resulting increase in computational and memory costs may prohibit the application of these MLIPs to perform large-scale MD simulations. Herein, we present a teacher-student training framework in which the latent knowledge from the teacher (atomic energies) is used to augment the students' training. We show that the light-weight student MLIPs have faster MD speeds at a fraction of the memory footprint compared to the teacher models. Remarkably, the student models can even surpass the accuracy of the teachers, even though both are trained on the same quantum chemistry dataset. Our work highlights a practical method for MLIPs to reduce the resources required for large-scale MD simulations.

36 MATERIALS SCIENCE↗
DOE OSTI2020

An Entropy-Maximization Approach to Automated Training Set Generation for Interatomic Potentials

Machine learning-based interatomic potentials are currently garnering a lot of attention as they strive to achieve the accuracy of electronic structure methods at the computational cost of empirical potentials. Given their generic functional forms, the transferability of these potentials is highly dependent on the quality of the training set, the generation of which can be highly labor-intensive. Good training sets should at once contain a very diverse set of configurations while avoiding redundancies that incur cost without providing benefits. We formalize these requirements in a local entropy-maximization framework and propose an automated sampling scheme to sample from this objective function. We show that this approach generates much more diverse training sets than unbiased sampling and is competitive with hand-crafted training sets.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2025

Empirically-calibrated H100 node power models for accurate AI training energy estimation

Accurately quantifying the energy use of artificial intelligence (AI) training is critical for infrastructure planning, carbon accounting, and sustainable data center operation, but few studies have directly measured the power consumption of production workloads on contemporary hardware. By combining empirical measurements from Brookhaven National Laboratory during AI training on 8-graphics-processing-unit H100 systems with open-source benchmarking data, we develop statistical models relating computational intensity to node-level power consumption. We measure the gap between manufacturer-rated thermal design power (TDP) and actual power demand during AI training. Our analysis reveals that even computationally intensive workloads operate at only 76% of the 10.2 kW TDP rating. Our architecture-specific model, calibrated to floating-point operations, predicts energy consumption with 11.4% mean absolute percentage error, significantly outperforming TDP-based approaches (27%–37% error). We identified distinct power signatures between transformer and convolutional neural network architectures, with transformers showing characteristic fluctuations that may impact grid stability. These results provide a measurement-grounded basis for improving AI training energy estimates, enabling more reliable infrastructure sizing, cost projections, and environmental impact assessments.

Newkirk, Alex C↗
DOE OSTI2021

Optimizing a magnitude-limited spectroscopic training sample for photometric classification of supernovae

ABSTRACT In preparation for photometric classification of transients from the Legacy Survey of Space and Time (LSST) we run tests with different training data sets. Using estimates of the depth to which the 4-m Multi-Object Spectroscopic Telescope (4MOST) Time Domain Extragalactic Survey (TiDES) can classify transients, we simulate a magnitude-limited sample reaching rAB ≈ 22.5 mag. We run our simulations with the software snmachine, a photometric classification pipeline using machine learning. The machine-learning algorithms struggle to classify supernovae when the training sample is magnitude limited, in contrast to representative training samples. Classification performance noticeably improves when we combine the magnitude-limited training sample with a simulated realistic sample of faint high-redshift supernovae observed from larger spectroscopic facilities; the algorithms’ range of average area under receiver operator characteristic curve (AUC) scores over 10 runs increases from 0.547–0.628 to 0.946–0.969 and purity of the classified sample reaches 95 per cent in all runs for two of the four algorithms. By creating new, artificial light curves using the augmentation software avocado, we achieve a purity in our classified sample of 95 per cent in all 10 runs performed for all machine-learning algorithms considered. We also reach a highest average AUC score of 0.986 with the artificial neural network algorithm. Having ‘true’ faint supernovae to complement our magnitude-limited sample is a crucial requirement in optimization of a 4MOST spectroscopic sample. However, our results are a proof of concept that augmentation is also necessary to achieve the best classification results.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Generation of shock trains in free liquid jets with a nanosecond green laser

Shock wave trains in liquid jets were previously generated only by ablation with femtosecond x-ray lasers. Here we show that shock trains in water microjets can be also generated using nanosecond green laser pulses with 1- to 10-mJ energy. Furthermore, the ablation of 15-, 20-, 30-, and 70-μm water microjets opened a gap in the jets and launched an initial shock wave. Fully developed shock trains were observed in the 30- and 70-μm jets up to 250-ns delays, and these trains were also transmitted inside the nozzles. A few tens of nanoseconds after the pulse, the shock dynamics and its pressure became similar to the ones generated by x-ray lasers, with a more rapid pressure decay in thinner jets. At time delays exceeding 100 ns in the 30-μm jets, the leading shock pressure stabilized to an approximately constant pressure of 40 MPa. The energy density deposited in the jets was estimated at 30 MJ/cm 3 by comparing the jet gaps in the green and x-ray laser experiments, and matched previous estimates for optical ablation in water. Here, the pressure decay in the 30-μm jets was modeled based on the pressure decay observed in x-ray laser experiments.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2025

Microstructural and rheological training and memory of nanocolloidal soft glasses under cyclic shear

An intrinsic feature of disordered and out-of-equilibrium materials, such as glasses, is the dependence of their properties on their history. An important example is rheological memory, in which disordered solids obtain properties based on their deformation history. Here, in this study, we employ x-ray photon correlation spectroscopy with in situ rheometry to characterize memory formation in a nanocolloidal soft glass due to cyclic shear. During a cycle, particles undergo irreversible displacements composed of a combination of shear-induced diffusion and heterogeneous, residual strain fields. At lower shear amplitudes, the displacements resemble a random walk in which the directions in each cycle are independent of those in preceding cycles, while at high amplitude, the irreversible displacements in consecutive cycles become correlated. The magnitudes of the displacements decrease with each cycle before reaching a steady state where the microstructure has been trained to achieve enhanced reversibility even at shear amplitudes well above yielding and despite the presence of thermal fluctuations. At amplitudes below and near yielding, these decreases are monotonic, while well above yielding, they are nonmonotonic, suggesting evidence of shear banding. Accompanying this microstructural training are corresponding decreases in the dissipation during each cycle and the magnitude of the residual stress toward steady-state values. Memory of the training is revealed by measurements in which the amplitude of the shear is changed after steady state is reached. The magnitude of the particle displacements, as well as the dissipation and the change in residual stress, vary nonmonotonically with the new shear amplitude, having minima near the training amplitude, thereby revealing correlated microscopic and macroscopic signatures of memory.

Chen, Yihao [Johns Hopkins Univ., Baltimore, MD (U↗
DOE OSTI2019

Single-Net Continual Learning with Progressive Segmented Training

There is an increasing need of continual learning in dynamic systems, such as the self-driving vehicle, the surveillance drone, and the robotic system. Such a system requires learning from the data stream, training the model to preserve previous information and adapt to a new task, and generating a single-headed vector for future inference. Different from previous approaches with dynamic structures, this work focuses on a single network and model segmentation to prevent catastrophic forgetting. Leveraging the redundant capacity of a single network, model parameters for each task are separated into two groups: one important group which is frozen to preserve current knowledge, and secondary group to be saved (not pruned) for a future learning. A fixed-size memory containing a small amount of previously seen data is further adopted to assist the training. Without additional regularization, the simple yet effective approach of Progressive Segmented Training (PST) successfully incorporates multiple tasks and achieves the state-of-the-art accuracy in the single-head evaluation on CIFAR-10 and CIFAR-100 datasets. Moreover, the segmented training significantly improves computation efficiency in continual learning at the edge.

Du, Xiaocong↗
DOE OSTI2021

Elastic distributed training with fast convergence and efficient resource utilization

Distributed learning is now routinely conducted on cloud as well as dedicated clusters. Training with elastic resources brings new challenges and design choices. Prior studies focus on runtime performance and assume a static algorithmic behavior. In this work, by analyzing the impact of of resource scaling on convergence, we introduce schedules for synchronous stochastic gradient descent that proactively adapt the number of learners to reduce training time and improve convergence. Our approach no longer assumes a constant number of processors throughout training. In our experiment, distributed stochastic gradient descent with dynamic schedules and reduction momentum achieves better convergence and significant speedups over prior static ones. Numerous distributed training jobs running on cloud may benefit from our approach.

Cong, Guojing↗
DOE OSTI2025

Accelerated Over-The-Air Neural Receiver Training Using Self-Contrastive Learning

Self-contrastive learning (SCL), a self-supervised learning method, has been shown to improve image and signal classifier accuracies and reduce the training time for neural communications receivers. In particular, prior work has shown that SCL applied as a pre-training step can improve simulated performance of OFDM in 3GPP TDL channel models by reducing the training time of the downstream classification task (demodulation and demapping). In this work a practical implementation demonstrating SCL pre-training using software defined radios (SDRs) is proposed.

Cooke, Corey [ORNL] (ORCID:0000000234263672)↗
DOE OSTI2023

Effect of CLIQ on Training of HL-LHC Quadrupole Magnets

The high-luminosity LHC upgrade requires stronger than LHC low-beta quadrupole magnets to reach the luminosity goals of the project. The project is well advanced and HL-LHC quadrupole magnets are currently being commissioned in US Labs (MQXFA magnets) and CERN (MQXFB magnets). Those are the first Nb 3 Sn magnets to be used in any large particle accelerator. At development stages, many Nb 3 Sn accelerator sub-scale models showed relatively slow training and MQXFA magnets were projected to have low tens of quenches before reaching operational field. Recently it was shown that dedicated capacitor-based de-vices can affect Nb 3 Sn magnet training, and it was suggested that CLIQ, a capacitor-based device intended for quench protection, can do too. Here, the present paper investigates effects on training likely induced by CLIQ, using the base fact that only half the coils in a quadrupole experience upward current modulation at quench because of capacitor discharge. The study encompasses all MQXFA production magnets trained at BNL to date. No other high-statistics data from identical magnets (series) with CLIQ protection exist so far. Implications and opportunities stemming from data analysis are discussed and conclusions drawn.

43 PARTICLE ACCELERATORS↗
DOE OSTI2018

SuperNeurons: Dynamic GPU Memory Management for Training Deep Neural Networks

Going deeper and wider in neural architectures improves their accuracy, while the limited GPU DRAM places an undesired restriction on the network design domain. Deep Learning (DL) practitioners either need to change to less de- sired network architectures, or nontrivially dissect a network across multiGPUs. These distract DL practitioners from concentrating on their original machine learning tasks. We present SuperNeurons: a dynamic GPU memory scheduling runtime to enable the network training far beyond the GPU DRAM capacity. SuperNeurons features 3 memory optimizations, Liveness Analysis, Unified Tensor Pool, and Cost-Aware Recomputation; together they effectively reduce the network-wide peak memory usage down to the maximal memory usage among layers. We also address the performance issues in these memory-saving techniques. Given the limited GPU DRAM, SuperNeurons not only provisions the necessary memory for the training, but also dynamically allocates the memory for convolution workspaces to achieve the high performance. Evaluations against Caffe, Torch, MXNet and TensorFlow have demonstrated that SuperNeurons trains at least 3.2432 deeper network than current ones with the leading performance. Particularly, SuperNeurons can train ResNet2500 that has 104 basic network layers on a 12GB K40c.

Wang, Linnan↗
DOE OSTI2025

Swap Path Network for Robust Person Search Pre-training

This code corresponds to the WACV25 conference paper, "Swap Path Network for Robust Person Search Pre-training". In that paper, we introduce a new model for the person search task called the Swap Path Net (SPNet). The person search task is a problem in computer vision, where we locate and rank matches to an image of a query person in a set of other images where we want to find them. We also introduce a novel pre-training algorithm specific to the Swap Path Net architecture. The code implements pre-training and fine-tuning of the Swap Path Net (SPNet). This includes ingesting image datasets and updating the weights of the SPNet neural network to train it for the person search task. The repository contains code, configs, and instructions to reproduce all results from the paper.

Jaffe, LucasW [Lawrence Livermore National Laborat↗
DOE OSTI2021

Skip-Connected Self-Recurrent Spiking Neural Networks with Joint Intrinsic Parameter and Synaptic Weight Training

As an important class of spiking neural networks (SNNs), recurrent spiking neural networks (RSNNs) possess great computational power and have been widely used for processing sequential data like audio and text. However, most RSNNs suffer from two problems. 1. Due to the lack of architectural guidance, random recurrent connectivity is often adopted, which does not guarantee good performance. 2. Training of RSNNs is in general challenging, bottlenecking achievable model accuracy. To address these problems, we propose a new type of RSNNs called Skip-Connected Self-Recurrent SNNs (ScSr-SNNs). Recurrence in ScSr-SNNs is introduced in a stereotyped manner by adding self-recurrent connections to spiking neurons. The SNNs with self-recurrent connections can realize recurrent behaviors similar to those of more complex RSNNs while the error gradients can be more straightforwardly calculated due to the mostly feedforward nature of the network. The network dynamics is enriched by skip connections between nonadjacent layers. Moreover, we propose a new backpropagation (BP) method called backpropagated intrinsic plasticity (BIP) to further boost the performance of ScSr-SNNs by training intrinsic model parameters. Unlike standard intrinsic plasticity rules that adjust the neuron's intrinsic parameters according to neuronal activity, the proposed BIP method optimizes intrinsic parameters based on the backpropagated error gradient of a well-defined global loss function in addition to synaptic weight training. Here, based on challenging speech, neuromorphic speech, and neuromorphic image datasets, the proposed ScSr-SNNs can boost performance by up to 2.85% compared with other types of RSNNs trained by state-of-the-art BP methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Adaptive language model training for molecular design

Abstract The vast size of chemical space necessitates computational approaches to automate and accelerate the design of molecular sequences to guide experimental efforts for drug discovery. Genetic algorithms provide a useful framework to incrementally generate molecules by applying mutations to known chemical structures. Recently, masked language models have been applied to automate the mutation process by leveraging large compound libraries to learn commonly occurring chemical sequences (i.e., using tokenization) and predict rearrangements (i.e., using mask prediction). Here, we consider how language models can be adapted to improve molecule generation for different optimization tasks. We use two different generation strategies for comparison, fixed and adaptive. The fixed strategy uses a pre-trained model to generate mutations; the adaptive strategy trains the language model on each new generation of molecules selected for target properties during optimization. Our results show that the adaptive strategy allows the language model to more closely fit the distribution of molecules in the population. Therefore, for enhanced fitness optimization, we suggest the use of the fixed strategy during an initial phase followed by the use of the adaptive strategy. We demonstrate the impact of adaptive training by searching for molecules that optimize both heuristic metrics, drug-likeness and synthesizability, as well as predicted protein binding affinity from a surrogate model. Our results show that the adaptive strategy provides a significant improvement in fitness optimization compared to the fixed pre-trained model, empowering the application of language models to molecular design tasks.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗