Search NASA⌕ Search

SEARCH · Search NASA

Results for “neural network 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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Reducing Communication in Graph Neural Network Training

Graph Neural Networks (GNNs) are powerful and flexible neural networks that use the naturally sparse connectivity information of the data. GNNs represent this connectivity as sparse matrices, which have lower arithmetic intensity and thus higher communication costs compared to dense matrices, making GNNs harder to scale to high concurrencies than convolutional or fully-connected neural networks. Here, we introduce a family of parallel algorithms for training GNNs and show that they can asymptotically reduce communication compared to previous parallel GNN training methods. We implement these algorithms, which are based on 1D, 1. 5D, 2D, and 3D sparse-dense matrix multiplication, using torch.distributed on GPU-equipped clusters. Our algorithms optimize communication across the full GNN training pipeline. We train GNNs on over a hundred GPUs on multiple datasets, including a protein network with over a billion edges.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Biologically-informed excitatory and inhibitory ratio for robust spiking neural network training

Spiking neural networks drawing inspiration from biological constraints of the brain promise an energy-efficient paradigm for artificial intelligence. However, challenges exist in identifying guiding principles to train these networks in a robust fashion. In addition, training becomes an even more difficult problem when incorporating biological constraints of excitatory and inhibitory connections. In this work, we identify several key factors, such as low initial firing rates and diverse inhibitory spiking patterns, that determine the overall ability to train in the context of spiking networks with various ratios of excitatory to inhibitory neurons. The results indicate networks with biologically-realistic excitatory:inhibitory ratios can reliably train at low activity levels and in noisy environments. Additionally, the Van Rossum distance, a measure of spike train synchrony, provides insight into the importance of inhibitory neurons to increase network robustness to noise. This work supports further biologically-informed large-scale networks and energy efficient hardware implementations.

bio-inspired computing↗
DOE OSTI2025

Development of systematic uncertainty-aware neural network trainings for binned-likelihood analyses at the LHC

We propose a neural network training method capable of accounting for the effects of systematic variations of the data model in the training process and describe its extension towards neural network multiclass classification. The procedure is evaluated on the realistic case of the measurement of Higgs boson production via gluon fusion and vector boson fusion in the τ τ decay channel at the CMS experiment. The neural network output functions are used to infer the signal strengths for inclusive production of Higgs bosons as well as for their production via gluon fusion and vector boson fusion. We observe improvements of 12 and 16% in the uncertainty in the signal strengths for gluon and vector-boson fusion, respectively, compared with a conventional neural network training based on cross-entropy.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

A Parallel Alternative for Energy-Efficient Neural Network Training and Inferencing

Energy efficiency of training and inferencing with large neural network models is a critical challenge facing the future of sustainable large-scale machine learning workloads. This paper introduces an alternative strategy, called phantom parallelism, to minimize the net energy consumption of traditional tensor (model) parallelism, the most energy-inefficient component of large neural network training. The approach is presented in the context of feed-forward network architectures as a preliminary, but comprehensive, proof-of-principle study of the proposed methodology. We derive new forward and backward propagation operators for phantom parallelism, implement them as custom autograd operations within an end-to-end phantom parallel training pipeline and compare its parallel performance and energy-efficiency against those of conventional tensor parallel training pipelines. Formal analyses that predict lower bandwidth and FLOP counts are presented with supporting empirical results on up to 256 GPUs that corroborate these gains. Experiments are shown to deliver ∼50% reduction in the energy consumed to train FFNs using the proposed phantom parallel approach when compared with conventional tensor parallel methods. Additionally, the proposed approach is shown to train smaller phantom models to the same model loss on smaller GPU counts as larger tensor parallel models on larger GPU counts offering the possibility for even greater energy savings.

Seal, Sudip [ORNL] (ORCID:0000000332330656)↗
DOE OSTI2023

Improving Deep Neural Networks’ Training for Image Classification With Nonlinear Conjugate Gradient-Style Adaptive Momentum

Momentum is crucial in stochastic gradient-based optimization algorithms for accelerating or improving training deep neural networks (DNNs). In deep learning practice, the momentum is usually weighted by a well-calibrated constant. However, tuning the hyperparameter for momentum can be a significant computational burden. In this article, we propose a novel adaptive momentum for improving DNNs training; this adaptive momentum, with no momentum-related hyperparame- ter required, is motivated by the nonlinear conjugate gradient (NCG) method. Stochastic gradient descent (SGD) with this new adaptive momentum eliminates the need for the momentum hyperparameter calibration, allows using a significantly larger learning rate, accelerates DNN training, and improves the final accuracy and robustness of the trained DNNs. For example, SGD with this adaptive momentum reduces classification errors for training ResNet110 for CIFAR10 and CIFAR100 from 5.25% to 4.64% and 23.75% to 20.03%, respectively. Furthermore, SGD, with the new adaptive momentum, also benefits adversarial training and, hence, improves the adversarial robustness of the trained DNNs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Training neural networks using Metropolis Monte Carlo and an adaptive variant

Abstract We examine the zero-temperature Metropolis Monte Carlo (MC) algorithm as a tool for training a neural network by minimizing a loss function. We find that, as expected on theoretical grounds and shown empirically by other authors, Metropolis MC can train a neural net with an accuracy comparable to that of gradient descent (GD), if not necessarily as quickly. The Metropolis algorithm does not fail automatically when the number of parameters of a neural network is large. It can fail when a neural network’s structure or neuron activations are strongly heterogenous, and we introduce an adaptive Monte Carlo algorithm (aMC) to overcome these limitations. The intrinsic stochasticity and numerical stability of the MC method allow aMC to train deep neural networks and recurrent neural networks in which the gradient is too small or too large to allow training by GD. MC methods offer a complement to gradient-based methods for training neural networks, allowing access to a distinct set of network architectures and principles.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Efficient I/O for Neural Network Training with Compressed Data

FanStore is a shared object store that enables efficient and scalable neural network training on supercomputers. By providing a global cache layer on node-local burst buffers using a compressed representation, it significantly enhances the processing capability of deep learning (DL) applications on existing hardware. In addition, FanStore allows POSIX-compliant file access to the compressed data in user space. We investigate the tradeoff between runtime overhead and data compression ratio using real-world datasets and applications, and propose a compressor selection algorithm to maximize storage capacity given performance constraints. We consider both asynchronous (i.e., with prefetching) and synchronous I/O strategies, and propose mechanisms for selecting compressors for both approaches. Using FanStore, the same storage hardware can host 2-13x more data for example applications without significant runtime overhead. Empirically, our experiments show that FanStore scales to 512 compute nodes with near linear performance scalability.

Zhang, Zhao↗
DOE OSTI2021

Confronting Domain Shift in Trained Neural Networks

SAND2021-15138 O This code accompanies a paper published in the Proceedings of Machine Learning Research (PMLR), “Confronting Domain Shift in Trained Neural Networks.” Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Martinez, Carianne↗
DOE OSTI2025

Convolutional Neural Networks Trained on Internal Variability Predict Forced Response of TOA Radiation by Learning the Pattern Effect

Abstract Predicting forced, long‐term radiative feedbacks from internal climate variability has been a decades‐long quest in climate science. We train a convolutional neural network (CNN) to predict annual‐ and global‐mean top of the atmosphere radiation anomalies from time‐varying maps of near‐surface temperature in climate models. Trained on internal variability alone, the nonlinear CNN can predict radiation under strong climate change, outperforms a regularized linear regression approach, and works within and across different climate models. We show with explainable artificial intelligence methods that the CNN draws predictive skill from physically meaningful regions but at much smaller spatial scales than currently assumed.

Rugenstein, Maria [Colorado State University Fort ↗
DOE OSTI2022

Do Neural Networks Trained with Topological Features Learn Different Internal Representations?

There is a growing body of work that leverages features extracted via topological data analysis to train machine learning models. While this field, sometimes known as topological machine learning (TML), has seen some notable successes, an understanding of how the process of learning from topological features differs from the process of learning from raw data is still limited. In this work, we begin to address one component of this larger issue by asking whether a model trained with topological features learns internal representations of data that are fundamentally different than those learned by a model trained with the original raw data. To quantify ``different'', we exploit two popular metrics that can be used to measure the similarity of the hidden representations of data within neural networks, neural stitching and centered kernel alignment. From these we draw a range of conclusions about how training with topological features does and does not change the representations that a model learns. Perhaps unsurprisingly, we find that structurally, the hidden representations of models trained and evaluated on topological features differ substantially compared to those trained and evaluated on the corresponding raw data. On the other hand, our experiments show that in some cases, these representations can be reconciled (at least to the degree required to solve the corresponding task) using a simple affine transformation. We conjecture that this means that neural networks trained on raw data may extract some limited topological features in the process of making predictions.

McGuire, Sarah L.↗
DOE OSTI2025

GSplit: Scaling Graph Neural Network Training on Large Graphs via Split-Parallelism

Graph neural networks (GNNs), an emerging class of machine learning models for graphs, have gained popularity for their superior performance in various graph analytical tasks. Mini-batch training is commonly used to train GNNs on large graphs, and data parallelism is the standard approach to scale mini-batch training across multiple GPUs. Data parallel approaches contain redundant work as subgraphs sampled by different GPUs contain significant overlap. To address this issue, we introduce a hybrid parallel mini-batch training paradigm called Split parallelism. Split parallelism avoids redundant work by splitting the sampling, loading, and training of each mini-batch across multiple GPUs. Split parallelism, however, introduces communication overheads that can be more than the savings from removing redundant work. We further present a lightweight partitioning algorithm that probabilistically minimizes these overheads. We implement spllit parllelism in GSplit and show that it outperforms state-of-the-art mini-batch training systems like DGL, Quiver, and P3.

Lim, Seung-Hwan [ORNL] (ORCID:0000000194616866)↗
DOE OSTI2020

PANTHER: A Programmable Architecture for Neural Network Training Harnessing Energy-Efficient ReRAM

The wide adoption of deep neural networks has been accompanied by ever-increasing energy and performance demands due to the expensive nature of training them. Additionally, numerous special-purpose architectures have been proposed to accelerate training: both digital and hybrid digital-analog using resistive RAM (ReRAM) crossbars. ReRAM-based accelerators have demonstrated the effectiveness of ReRAM crossbars at performing matrix-vector multiplication operations that are prevalent in training. However, they still suffer from inefficiency due to the use of serial reads and writes for performing the weight gradient and update step. A few works have demonstrated the possibility of performing outer products in crossbars, which can be used to realize the weight gradient and update step without the use of serial reads and writes. However, these works have been limited to low precision operations which are not sufficient for typical training workloads. Moreover, they have been confined to a limited set of training algorithms for fully-connected layers only. To address these limitations, we propose a bit-slicing technique for enhancing the precision of ReRAM-based outer products, which is substantially different from bit-slicing for matrix-vector multiplication only. We incorporate this technique into a crossbar architecture with three variants catered to different training algorithms. To evaluate our design on different types of layers in neural networks (fully-connected, convolutional, etc.) and training algorithms, we develop PANTHER, an ISA-programmable training accelerator with compiler support. Our design can also be integrated into other accelerators in the literature to enhance their efficiency. Our evaluation shows that PANTHER achieves up to 8.02×, 54.21×, and 103× energy reductions as well as 7.16×, 4.02×, and 16× execution time reductions compared to digital accelerators, ReRAM-based accelerators, and GPUs, respectively.

42 ENGINEERING↗
DOE OSTI2023

NSGA-PINN: A Multi-Objective Optimization Method for Physics-Informed Neural Network Training

This paper presents NSGA-PINN, a multi-objective optimization framework for the effective training of physics-informed neural networks (PINNs). The proposed framework uses the non-dominated sorting genetic algorithm (NSGA-II) to enable traditional stochastic gradient optimization algorithms (e.g., ADAM) to escape local minima effectively. Additionally, the NSGA-II algorithm enables satisfying the initial and boundary conditions encoded into the loss function during physics-informed training precisely. We demonstrate the effectiveness of our framework by applying NSGA-PINN to several ordinary and partial differential equation problems. In particular, we show that the proposed framework can handle challenging inverse problems with noisy data.

Lu, Binghang (ORCID:0009000160016632)↗
DOE OSTI2024

MDLoader: A Hybrid Model-Driven Data Loader for Distributed Graph Neural Network Training

Scalable data management is essential for processing large scientific dataset on HPC platforms for distributed deep learning. In-memory distributed storage is preferred for its speed, enabling rapid, random, and frequent data access required by stochastic optimizers. Processes use one-sided or collective communication to fetch remote data, with optimal performance depending on (i) dataset characteristics, (ii) training scale, and (iii) interconnection network. Empirical analysis shows collective communication excels with larger mini-batch sizes and/or fewer processes, whereas one-sided communication outperforms at larger scales. We propose MDLoader, a hybrid in-memory data loader for distributed graph neural network training. MDLoader features a model-driven performance estimator that dynamically selects between one-sided and collective communication at the beginning of training using Tree of Parzen Estimators (TPE). Evaluations on NERSC Perlmutter and OLCF Summit show MDLoader outperforms single-backend loaders by up to 2.83 × and predicts the suitable communication method with 96.3% (Perlmutter) and 94.3% (Summit) success rate.

Bae, Jonghyun↗
DOE OSTI2023

Runtime extension for neural network training with heterogeneous memory

Systems, apparatuses, and methods for managing buffers in a neural network implementation with heterogeneous memory are disclosed. A system includes a neural network coupled to a first memory and a second memory. The first memory is a relatively low-capacity, high-bandwidth memory while the second memory is a relatively high-capacity, low-bandwidth memory. During a forward propagation pass of the neural network, a run-time manager monitors the usage of the buffers for the various layers of the neural network. During a backward propagation pass of the neural network, the run-time manager determines how to move the buffers between the first and second memories based on the monitored buffer usage during the forward propagation pass. As a result, the run-time manager is able to reduce memory access latency for the layers of the neural network during the backward propagation pass.

Mappouras, Georgios↗
DOE OSTI2021

Confronting Domain Shift in Trained Neural Networks

Neural networks (NNs) are known as universal function approximators and can interpolate nonlinear functions between observed data points. However, when the target domain for deployment shifts from the training domain and NNs must extrapolate, the results are notoriously poor. Prior work Martinez et al. (2019) has shown that NN uncertainty estimates can be used to correct binary predictions in shifted domains without retraining the model. We hypothesize that this approach can be extended to correct real-valued time series predictions. As an exemplar, we consider two mechanical systems with nonlinear dynamics. The first system consists of a spring-mass system where the stiffness changes abruptly, and the second is a real experimental system with a frictional joint that is an open challenge for structural dynamicists to model efficiently. Our experiments will test whether 1) NN uncertainty estimates can identify when the input domain has shifted from the training domain and 2) whether the information used to calculate uncertainty estimates can be used to correct the NN’s time series predictions. While the method as proposed did not significantly improve predictions, our results did show potential for modifications that could improve models’ predictions and play a role in structural health monitoring systems that directly impact public safety.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Enhancing Docking Accuracy with PECAN2, a 3D Atomic Neural Network Trained without Co-Complex Crystal Structures

Decades of drug development research have explored a vast chemical space for highly active compounds. The exponential growth of virtual libraries enables easy access to billions of synthesizable molecules. Computational modeling, particularly molecular docking, utilizes physics-based calculations to prioritize molecules for synthesis and testing. Nevertheless, the molecular docking process often yields docking poses with favorable scores that prove to be inaccurate with experimental testing. To address these issues, several approaches using machine learning (ML) have been proposed to filter incorrect poses based on the crystal structures. However, most of the methods are limited by the availability of structure data. Here, we propose a new pose classification approach, PECAN2 (Pose Classification with 3D Atomic Network 2), without the need for crystal structures, based on a 3D atomic neural network with Point Cloud Network (PCN). The new approach uses the correlation between docking scores and experimental data to assign labels, instead of relying on the crystal structures. We validate the proposed classifier on multiple datasets including human mu, delta, and kappa opioid receptors and SARS-CoV-2 Mpro. Our results demonstrate that leveraging the correlation between docking scores and experimental data alone enhances molecular docking performance by filtering out false positives and false negatives.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗