Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Neural Network”

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 397 records · Page 22

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↗

Optimized asynchronous training of neural networks using a distributed parameter server with eager updates

A method of training a neural network includes, at a local computing node, receiving remote parameters from a set of one or more remote computing nodes, initiating execution of a forward pass in a local neural network in the local computing node to determine a final output based on the remote parameters, initiating execution of a backward pass in the local neural network to determine updated parameters for the local neural network, and prior to completion of the backward pass, transmitting a subset of the updated parameters to the set of remote computing nodes.

Hamidouche, Khaled↗

Keras2c: A library for converting Keras neural networks to real-time compatible C

With the growth of machine learning models and neural networks in measurement and control systems comes the need to deploy these models in a way that is compatible with existing systems. Existing options for deploying neural networks either introduce very high latency, require expensive and time consuming work to integrate into existing code bases, or only support a very limited subset of model types. We have therefore developed a new method called Keras2c, which is a simple library for converting Keras/TensorFlow neural network models into real-time compatible C code. It supports a wide range of Keras layers and model types including multidimensional convolutions, recurrent layers, multi-input/output models, and shared layers. Keras2c re-implements the core components of Keras/TensorFlow required for predictive forward passes through neural networks in pure C, relying only on standard library functions considered safe for real-time use. The core functionality consists of ~ 1500 lines of code, making it lightweight and easy to integrate into existing codebases. Keras2c has been successfully tested in experiments and is currently in use on the plasma control system at the DIII-D National Fusion Facility at General Atomics in San Diego.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

The Efficacy of Galaxy Shape Parameters in Photometric Redshift Estimation: A Neural Network Approach

We present a determination of the effects of including galaxy morphological parameters in photometric redshift estimation with an artificial neural network method. Neural networks, which recognize patterns in the information content of data in an unbiased way, can be a useful estimator of the additional information contained in extra parameters, such as those describing morphology, if the input data are treated on an equal footing. We show that certain principal components of the morphology information are correlated with galaxy type. However, we find that for the data used the inclusion of morphological information does not have a statistically significant benefit for photometric redshift estimation with the techniques employed here. The inclusion of these parameters may result in a trade-off between extra information and additional noise, with the additional noise becoming more dominant as more parameters are added.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Hideen Markov Models and Neural Networks for Fault Detection in Dynamic Systems

None given. (From conclusion): Neural networks plus Hidden Markov Models(HMM)can provide excellene detection and false alarm rate performance in fault detection applications. Modified models allow for novelty detection. Also covers some key contributions of neural network model, and application status.

space network↗

Multiverse: Bayesian model selection for neural networks

Multiverse is a code repository for a set of tools for Bayesian model selection for neural networks. The goal of the tools is to provide capabilities for selecting among prior and model specifications for Bayesian neural networks. Multiverse will include tools for creating Bayesian neural networks, evaluating the Bayesian model evidence, and performing inference in Bayesian neural networks. These components are written in Python, a high-level programming language that takes advantage of the Python ecosystem of high-quality open-source packages for machine learning and signal processing.

Klein, Natalie↗

Evaluating the performance of two neutron spectrum unfolding codes based on iterative procedures and artificial neural networks

In this work the performance of two neutron spectrum unfolding codes based on iterative procedures and artificial neural networks is evaluated. The first one code based on traditional iterative procedures and called Neutron spectrometry and dosimetry from the Universidad Autonoma de Zacatecas (NSDUAZ) use the SPUNIT iterative algorithm and was designed to unfold neutron spectrum and calculate 15 dosimetric quantities and 7 IAEA survey meters. The main feature of this code is the automated selection of the initial guess spectrum trough a compendium of neutron spectrum compiled by the IAEA. The second one code known as Neutron spectrometry and dosimetry with artificial neural networks (NDSann) is a code designed using neural nets technology. The artificial intelligence approach of neural net does not solve mathematical equations. By using the knowledge stored at synaptic weights on a neural net properly trained, the code is capable to unfold neutron spectrum and to simultaneously calculate 15 dosimetric quantities, needing as entrance data, only the rate counts measured with a Bonner spheres system. Similarities of both NSDUAZ and NSDann codes are: they follow the same easy and intuitive user's philosophy and were designed in a graphical interface under the LabVIEW programming environment. Both codes unfold the neutron spectrum expressed in 60 energy bins, calculate 15 dosimetric quantities and generate a full report in HTML format. Differences of these codes are: NSDUAZ code was designed using classical iterative approaches and needs an initial guess spectrum in order to initiate the iterative procedure. In NSDUAZ, a programming routine was designed to calculate 7 IAEA instrument survey meters using the fluence-dose conversion coefficients. NSDann code use artificial neural networks for solving the ill-conditioned equation system of neutron spectrometry problem through synaptic weights of a properly trained neural network. Contrary to iterative procedures, in neural net approach it is possible to reduce the rate counts used to unfold the neutron spectrum. To evaluate these codes a computer tool called Neutron Spectrometry and dosimetry computer tool was designed. The results obtained with this package are showed. The codes here mentioned are freely available upon request to the authors.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Training product unit neural networks with genetic algorithms

The training of product neural networks using genetic algorithms is discussed. Two unusual neural network techniques are combined; product units are employed instead of the traditional summing units and genetic algorithms train the network rather than backpropagation. As an example, a neural netork is trained to calculate the optimum width of transistors in a CMOS switch. It is shown how local minima affect the performance of a genetic algorithm, and one method of overcoming this is presented.

Janson, D. J.↗

The application of neural networks to the SSME startup transient

Feedforward neural networks were used to model three parameters during the Space Shuttle Main Engine startup transient. The three parameters were the main combustion chamber pressure, a controlled parameter, the high pressure oxidizer turbine discharge temperature, a redlined parameter, and the high pressure fuel pump discharge pressure, a failure-indicating performance parameter. Network inputs consisted of time windows of data from engine measurements that correlated highly to the modeled parameter. A standard backpropagation algorithm was used to train the feedforward networks on two nominal firings. Each trained network was validated with four additional nominal firings. For all three parameters, the neural networks were able to accurately predict the data in the validation sets as well as the training set.

Meyer, Claudia M.↗

The application of neural networks to the SSME startup transient

Feedforward neural networks were used to model three parameters during the Space Shuttle Main Engine startup transient. The three parameters were the main combustion chamber pressure, a controlled parameter, the high pressure oxidizer turbine discharge temperature, a redlined parameter, and the high pressure fuel pump discharge pressure, a failure-indicating performance parameter. Network inputs consisted of time windows of data from engine measurements that correlated highly to the modeled parameter. A standard backpropagation algorithm was used to train the feedforward networks on two nominal firings. Each trained network was validated with four additional nominal firings. For all three parameters, the neural networks were able to accurately predict the data in the validation sets as well as the training set.

Meyer, Claudia M.↗

A decade of neural networks: Practical applications and prospects

On May 11-13, 1994, JPL's Center for Space Microelectronics Technology (CSMT) hosted a neural network workshop entitled, 'A Decade of Neural Networks: Practical Applications and Prospects,' sponsored by DOD and NASA. The past ten years of renewed activity in neural network research has brought the technology to a crossroads regarding the overall scope of its future practical applicability. The purpose of the workshop was to bring together the sponsoring agencies, active researchers, and the user community to formulate a vision for the next decade of neural network research and development prospects, with emphasis on practical applications. Of the 93 participants, roughly 15% were from government agencies, 30% were from industry, 20% were from universities, and 35% were from Federally Funded Research and Development Centers (FFRDC's).

Kemeny, Sabrina↗

Attitude control of spacecraft using neural networks

This paper investigates the use of radial basis function neural networks for adaptive attitude control and momentum management of spacecraft. In the first part of the paper, neural networks are trained to learn from a family of open-loop optimal controls parameterized by the initial states and times-to-go. The trained is then used for closed-loop control. In the second part of the paper, neural networks are used for direct adaptive control in the presence of unmodeled effects and parameter uncertainty. The control and learning laws are derived using the method of Lyapunov.

Vadali, Srinivas R.↗

Neural network application to comprehensive engine diagnostics

We have previously reported on the use of neural networks for detection and identification of faults in complex microprocessor controlled powertrain systems. The data analyzed in those studies consisted of the full spectrum of signals passing between the engine and the real-time microprocessor controller. The specific task of the classification system was to classify system operation as nominal or abnormal and to identify the fault present. The primary concern in earlier work was the identification of faults, in sensors or actuators in the powertrain system as it was exercised over its full operating range. The use of data from a variety of sources, each contributing some potentially useful information to the classification task, is commonly referred to as sensor fusion and typifies the type of problems successfully addressed using neural networks. In this work we explore the application of neural networks to a different diagnostic problem, the diagnosis of faults in newly manufactured engines and the utility of neural networks for process control.

Marko, Kenneth A.↗

Modeling design and control problems involving neural network surrogates

Here, we consider nonlinear optimization problems that involve surrogate models represented by neural networks. We demonstrate first how to directly embed neural network evaluation into optimization models, highlight a difficulty with this approach that can prevent convergence, and then characterize stationarity of such models. We then present two alternative formulations of these problems in the specific case of feedforward neural networks with ReLU activation: as a mixed-integer optimization problem and as a mathematical program with complementarity constraints. For the latter formulation we prove that stationarity at a point for this problem corresponds to stationarity of the embedded formulation. Each of these formulations may be solved with state-of-the-art optimization methods, and we show how to obtain good initial feasible solutions for these methods. We compare our formulations on three practical applications arising in the design and control of combustion engines, in the generation of adversarial attacks on classifier networks, and in the determination of optimal flows in an oil well network.

97 MATHEMATICS AND COMPUTING↗

A Classification of G -invariant Shallow Neural Networks

When trying to fit a deep neural network (DNN) to a G-invariant target function with G a group, it only makes sense to constrain the DNN to be G-invariant as well. However, there can be many different ways to do this, thus raising the problem of “G-invariant neural architecture design”: What is the optimal Ginvariant architecture for a given problem? Before we can consider the optimization problem itself, we must understand the search space, the architectures in it, and how they relate to one another. In this paper, we take a first step towards this goal; we prove a theorem that gives a classification of all G-invariant single-hidden-layer or “shallow” neural network (G-SNN) architectures with ReLU activation for any finite orthogonal group G, and we prove a second theorem that characterizes the inclusion maps or “network morphisms” between the architectures that can be leveraged during neural architecture search (NAS). The proof is based on a correspondence of every G-SNN to a signed permutation representation of G acting on the hidden neurons; the classification is equivalently given in terms of the first cohomology classes of G, thus admitting a topological interpretation. The G-SNN architectures corresponding to nontrivial cohomology classes have, to our knowledge, never been explicitly identified in the literature previously. Using a code implementation, we enumerate the G-SNN architectures for some example groups G and visualize their structure. Lastly, we prove that architectures corresponding to inequivalent cohomology classes coincide in function space only when their weight matrices are zero, and we discuss the implications of this for NAS.

Agrawal, Devanshu↗

Graph Neural Networks and Applied Linear Algebra v.1.0

SAND2024-01365O The Graph Neural Networks and Applied Linear Algebra is companion software for the educational article with the same title. The software provides illustrative examples of graph neural networks in Matlab and Python. These stand-alone algorithms are for educational purposes. The software also includes graph neural network-based algorithms for a trainable Jacobi iteration as well as diffusion coefficient estimation. The software provides human-interpretable implementations of Graph Neural Networks in Matlab and Python. These implementations are not optimized for performance and instead emphasize readability. 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.

SciDAC↗

Polyconvex neural network models of thermoelasticity

Machine-learning function representations such as neural networks have proven to be excellent constructs for constitutive modeling due to their flexibility to represent highly nonlinear data and their ability to incorporate constitutive constraints, which also allows them to generalize well to unseen data. Here, in this work, we extend a polyconvex hyperelastic neural network framework to (isotropic) thermo-hyperelasticity by specifying the thermodynamic and material theoretic requirements for an expansion of the Helmholtz free energy expressed in terms of deformation invariants and temperature. Different formulations which a priori ensure polyconvexity with respect to deformation and concavity with respect to temperature are proposed and discussed. The physics-augmented neural networks are furthermore calibrated with a recently proposed sparsification algorithm that not only aims to fit the training data but also penalizes the number of active parameters, which prevents overfitting in the low data regime and promotes generalization. The performance of the proposed framework is demonstrated on synthetic data, which illustrate the expected thermomechanical phenomena, and existing temperature-dependent uniaxial tension and tension-torsion experimental datasets.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Evolving Ensembles of Spiking Neural Networks for Neuromorphic Systems

Evolutionary algorithms have been proposed as a solution to overcome many of the challenges associated with training spiking neural networks. While evolutionary optimization for spiking neural networks is very flexible, its performance has difficulty scaling to complex tasks and correspondingly complex network structures. Here we propose a method for evolving ensembles of spiking neural networks. By using ensemble learning, the flexibility of evolutionary optimization is fully preserved while scaling to more challenging tasks. We test the performance of the proposed method using handwritten digit classification. We investigate multiple strategies for constructing ensembles of spiking neural networks, and demonstrate that evolving ensembles of SNNs offers significant performance advantages over evolutionary optimization.

Elbrecht, Daniel↗