Search NASA⌕ Search

SEARCH · Search NASA

Results for “Neural network compression”

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 145 records · Page 8

Predicting concrete compressive strength using hybrid ensembling of surrogate machine learning models

This study aims to implement a hybrid ensemble surrogate machine learning technique in predicting the compressive strength (CS) of concrete, an important parameter used for durability design and service life prediction of concrete structures in civil engineering projects. For this purpose, an experimental database consisting of 1030 records has been compiled from the machine learning repository of the University of California, Irvine. The database was used to train and validate four conventional machine learning (CML) models, namely Artificial Neural Network (ANN), Linear and Non-Linear Multivariate Adaptive Regression Splines (MARS-L and MARS-C), Gaussian Process Regression (GPR), and Minimax Probability Machine Regression (MPMR). Subsequently, the predicted outputs of CML models were combined and trained using ANN to construct the Hybrid Ensemble Model (HENSM). It is observed that the proposed HENSM produces higher predictive accuracy compared to the CML models used in the present study. The predictive performance of all models for CS prediction was compared using the testing dataset and it is found that the HENSM model attained the highest predictive accuracy in both phases. Based on the experimental results, the newly constructed HENSM model is very potential to be a new alternative in handling the overfitting issues of CML models and hence, can be used to predict the concrete CS, including the design of less polluting and more sustainable concrete constructions.

36 MATERIALS SCIENCE↗

Visualization Quality Assessment

Understanding how inaccuracies in visualizations affect users’ perception and understanding of scientific data is hard. Inaccuracies in visualizations are quite common and could arise from a range of sources such as errors in the original dataset arising from compression artifacts, errors in the capturing device, noise during transmission of the data, effects due to the algorithm being used to convert data to visualization images, images generated from neural networks, and sources we have yet to discover. Many image quality assessment metrics have been developed to quantify image errors. However, these are usually focused on “natural images” rather than visualizations of scientific data. Common image quality assessment metrics (IQAs) include MSE, PSNR, perceptual metrics such SSIM, FSIM as well as perceptual metrics using deep learning approaches. However, a critical part of understanding how errors are perceived by humans, and subsequently developing more accurate quality assessment metrics, is through user evaluation studies. The goal of this software is to develop a visualization quality assessment (VQA) process that will enable the generation of VQAs that can be used to quantify errors in scientific data visualizations. The VQA development process will include software to support user evaluation experimental design, analysis of visualization differences against standard quality metrics, and the ability to develop additional VQA metrics specific to scientific visualization images.

Grosset, Andre↗

Data-Efficient Dimensionality Reduction and Surrogate Modeling of High-Dimensional Stress Fields

Tensor datatypes representing field variables like stress, displacement, velocity, etc., have increasingly become a common occurrence in data-driven modeling and analysis of simulations. Numerous methods [such as convolutional neural networks (CNNs)] exist to address the meta-modeling of field data from simulations. As the complexity of the simulation increases, so does the cost of acquisition, leading to limited data scenarios. Modeling of tensor datatypes under limited data scenarios remains a hindrance for engineering applications. Here, in this article, we introduce a direct image-to-image modeling framework of convolutional autoencoders enhanced by information bottleneck loss function to tackle the tensor data types with limited data. The information bottleneck method penalizes the nuisance information in the latent space while maximizing relevant information making it robust for limited data scenarios. The entire neural network framework is further combined with robust hyperparameter optimization. We perform numerical studies to compare the predictive performance of the proposed method with a dimensionality reduction-based surrogate modeling framework on a representative linear elastic ellipsoidal void problem with uniaxial loading. The data structure focuses on the low-data regime (fewer than 100 data points) and includes the parameterized geometry of the ellipsoidal void as the input and the predicted stress field as the output. The results of the numerical studies show that the information bottleneck approach yields improved overall accuracy and more precise prediction of the extremes of the stress field. Additionally, an in-depth analysis is carried out to elucidate the information compression behavior of the proposed framework.

artificial intelligence↗

Integrated photonic encoder for low power and high-speed image processing

Abstract Modern lens designs are capable of resolving greater than 10 gigapixels, while advances in camera frame-rate and hyperspectral imaging have made data acquisition rates of Terapixel/second a real possibility. The main bottlenecks preventing such high data-rate systems are power consumption and data storage. In this work, we show that analog photonic encoders could address this challenge, enabling high-speed image compression using orders-of-magnitude lower power than digital electronics. Our approach relies on a silicon-photonics front-end to compress raw image data, foregoing energy-intensive image conditioning and reducing data storage requirements. The compression scheme uses a passive disordered photonic structure to perform kernel-type random projections of the raw image data with minimal power consumption and low latency. A back-end neural network can then reconstruct the original images with structural similarity exceeding 90%. This scheme has the potential to process data streams exceeding Terapixel/second using less than 100 fJ/pixel, providing a path to ultra-high-resolution data and image acquisition systems.

47 OTHER INSTRUMENTATION↗

Normalizing flows for high-dimensional detector simulations

Whenever invertible generative networks are needed for LHC physics, normalizing flows show excellent performance. In this work, we investigate their performance for fast calorimeter shower simulations with increasing phase space dimension. We use fast and expressive coupling spline transformations applied to the CaloChallenge datasets. In addition to the base flow architecture we also employ a VAE to compress the dimensionality and train a generative network in the latent space. We evaluate our networks on several metrics, including high-level features, classifiers, and generation timing. Our findings demonstrate that invertible neural networks have competitive performance when compared to autoregressive flows, while being substantially faster during generation.

Ernst, Florian↗

Probabilisitc Geobiological Classification Using Elemental Abundance Distributions and Lossless Image Compression in Recent and Modern Organisms

Last year we presented techniques for the detection of fossils during robotic missions to Mars using both structural and chemical signatures[Storrie-Lombardi and Hoover, 2004]. Analyses included lossless compression of photographic images to estimate the relative complexity of a putative fossil compared to the rock matrix [Corsetti and Storrie-Lombardi, 2003] and elemental abundance distributions to provide mineralogical classification of the rock matrix [Storrie-Lombardi and Fisk, 2004]. We presented a classification strategy employing two exploratory classification algorithms (Principal Component Analysis and Hierarchical Cluster Analysis) and non-linear stochastic neural network to produce a Bayesian estimate of classification accuracy. We now present an extension of our previous experiments exploring putative fossil forms morphologically resembling cyanobacteria discovered in the Orgueil meteorite. Elemental abundances (C6, N7, O8, Na11, Mg12, Ai13, Si14, P15, S16, Cl17, K19, Ca20, Fe26) obtained for both extant cyanobacteria and fossil trilobites produce signatures readily distinguishing them from meteorite targets. When compared to elemental abundance signatures for extant cyanobacteria Orgueil structures exhibit decreased abundances for C6, N7, Na11, All3, P15, Cl17, K19, Ca20 and increases in Mg12, S16, Fe26. Diatoms and silicified portions of cyanobacterial sheaths exhibiting high levels of silicon and correspondingly low levels of carbon cluster more closely with terrestrial fossils than with extant cyanobacteria. Compression indices verify that variations in random and redundant textural patterns between perceived forms and the background matrix contribute significantly to morphological visual identification. The results provide a quantitative probabilistic methodology for discriminating putatitive fossils from the surrounding rock matrix and &om extant organisms using both structural and chemical information. The techniques described appear applicable to the geobiological analysis of meteoritic samples or in situ exploration of the Mars regolith. Keywords: cyanobacteria, microfossils, Mars, elemental abundances, complexity analysis, multifactor analysis, principal component analysis, hierarchical cluster analysis, artificial neural networks, paleo-biosignatures

Storrie-Lombardi, Michael C.↗

Detection of Synchrophasor False Data Injection Attack using Feature Interactive Network

The synchrophasor data recorded by Phasor Measurement Units (PMUs) plays an increasingly critical role in the regulation and situational awareness of power systems. However, the widely installed PMUs are vulnerable to multiple malicious attacks from cyber hackers during data transmission and storage. To address this problem, a Modified Ensemble Empirical Mode Decomposition (MEEMD) is proposed first to extract the intrinsic mode functions of each Synchrophasor Data Attacks (SDA). The frequency-based adaptive screening criterion embedded in MEEMD is used to eliminate the false intrinsic mode functions. Next, a Multivariate Convolutional Neural Network (MCNN) is proposed to identify multiple SDA by utilizing the extracted intrinsic mode functions and original SDA as input vectors. A fusion block as the main structure of MCNN is also leveraged to increase the diversity of features and compress the model parameters. Integrating MEEMD and MCNN, a framework with automatic feature extraction and multi-source information fusion capability, referred to as Feature Interactive Network (FIN), is proposed to detect multiple SDA. Based on the proposed FIN framework, six types of SDA are explored for the first time using actual synchrophasor data in FNET/Grideye that was collected from different locations in the U.S. Eastern Interconnection. Finally, a large quantity of experiments with different attack strengths are used to evaluate the adaptability and classification performance of the proposed FIN.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Probabilistic Classification Using Elemental Abundance Distributions and Lossless Image Compression in Apollo 17 Lunar Dust Samples from Mare Serenitatis

We have previously outlined a strategy for the detection of fossils [Storrie-Lombardi and Hoover, 2004] and extant microbial life [Storrie-Lombaudi and Hoover, 20051 during robotic missions to Mars using co-registered structural and chemical signatures. Data inputs included image lossless compression indices to estimate relative textural complexity and elemental abundance distributions. Two exploratory classification algorithms (principal component analysis and hierarchical cluster analysis) provide an initial tentative classification of all targets. Nonlinear stochastic neural networks are then trained to produce a Bayesian estimate of algorithm classification accuracy. The strategy previously has been successful in distinguishing regions of biotic and abiotic alteration of basalt glass from unaltered samples. [Storrie-Lombardi and Fisk, 2004; Storrie-Lombardi and Fisk, 2004] Such investigations of abiotic versus biotic alteration of terrestrial mineralogy on Earth are compromised by .the difficulty finding mineralogy completely unaffected by the ubiquitous presence of microbial life on the planet. The renewed interest in lunar exploration offers an opportunity to investigate geological materials that may exhibit signs of aqueous alteration, but are highly unlikely to contain contaminating biological weathering signatures. We here present an extension of our earlier data set to include lunar dust samples obtained during the Apollo 17 mission. Apollo 17 landed in the Taurus-Littrow Valley in Mare Serenitatis. Most of the rock samples from this region of the lunar highlands are basalts comprised primarily of plagioclase and pyroxene and selected examples of orange and black volcanic glass. SEM images and elemental abundances (C6, N7, O8, Na11, Mg12, Al13, Si14, P15, S16, Cll7, K19, Ca20, Fe26) for a series of targets in the lunar dust samples are compared to the extant cyanobacteria, fossil trilobites, Orgueil meteorite, and terrestrial basalt targets previously discussed. The data set provides a first step in producing a quantitative probabilistic methodology for geobiological analysis of returned lunar samples or in situ exploration.

Storrie-Lombardi, Michael C.↗

Constitutive Modeling of High-Temperature Flow Stress of Armor Steel in Ballistic Applications: A Comparative Study

Armox 500T is one of the armor grade steels extensively used as armor against bullet penetration in ballistic applications. In such applications, the material undergoes plastic deformation at large strain rates (10{sup 3} s{sup −1}) at temperatures of the order of 673-1373 K. In the present work, an attempt has been made to predict strain rate and temperature-dependent flow behavior of Armox 500T steel through physical-based model like modified Zerilli–Armstrong (M-ZA) and phenomenological-based models like Cowper Symonds (CS), modified Johnson–Cook (M-JC), Arrhenius (Arr.) and Khan–Huang–Liang (KHL) constitutive material model. Isothermal uniaxial compression tests at low strain rates (10{sup −3}-10{sup −1} s{sup −1}) and dynamic compression tests at high strain rates (600-3000 s{sup −1}) in the temperatures range of 673-1373 K have been carried out to determine constitutive material model parameters. In addition, an artificial neural network (ANN) model that works on multilayer perceptron (MLP) based back propagation neural network (BPNN) has also been developed. The results from all these models have been compared in terms of three statistical parameters namely correlation coefficient (R), mean absolute percent error (MAPE) and its standard deviation (Δ). The results revealed that M-JC model shows highest correlation coefficient and lowest average absolute error. Further, ANN model prediction has the highest accuracy with R = 0.999 and MAPE = 1.052.

36 MATERIALS SCIENCE↗

DeepAdversaries: examining the robustness of deep learning models for galaxy morphology classification

With increased adoption of supervised deep learning methods for work with cosmological survey data, the assessment of data perturbation effects (that can naturally occur in the data processing and analysis pipelines) and the development of methods that increase model robustness are increasingly important. In the context of morphological classification of galaxies, we study the effects of perturbations in imaging data. In particular, we examine the consequences of using neural networks when training on baseline data and testing on perturbed data. We consider perturbations associated with two primary sources: (a) increased observational noise as represented by higher levels of Poisson noise and (b) data processing noise incurred by steps such as image compression or telescope errors as represented by one-pixel adversarial attacks. We also test the efficacy of domain adaptation techniques in mitigating the perturbation-driven errors. We use classification accuracy, latent space visualizations, and latent space distance to assess model robustness in the face of these perturbations. For deep learning models without domain adaptation, we find that processing pixel-level errors easily flip the classification into an incorrect class and that higher observational noise makes the model trained on low-noise data unable to classify galaxy morphologies. On the other hand, we show that training with domain adaptation improves model robustness and mitigates the effects of these perturbations, improving the classification accuracy up to 23% on data with higher observational noise. Domain adaptation also increases up to a factor of ${\approx}2.3$ the latent space distance between the baseline and the incorrectly classified one-pixel perturbed image, making the model more robust to inadvertent perturbations. Successful development and implementation of methods that increase model robustness in astronomical survey pipelines will help pave the way for many more uses of deep learning for astronomy.

79 ASTRONOMY AND ASTROPHYSICS↗

Ephemeral Learning - Augmenting Triggers with Online-Trained Normalizing Flows

The large data rates at the LHC require an online trigger system to select relevant collisions. Rather than compressing individual events, we propose to compress an entire data set at once. We use a normalizing flow as a deep generative model to learn the probability density of the data online. The events are then represented by the generative neural network and can be inspected offline for anomalies or used for other analysis purposes. We demonstrate our new approach for a toy model and a correlation-enhanced bump hunt.

97 MATHEMATICS AND COMPUTING↗

Vector Quantization With Emergent Codebook Structure

Proposed scheme under development for transmission of vector-quantized digital video images, vector quantizer codebook updated to adapt quantizer to changing signal statistics. Intended to be realized with electronic neural network. Codebook, which consists of patterns constituting video images, will undergo training during operation and scheme will develop codebooks ordered during training. System enables coding more compact, more immune to noise, and supports variable rate compression.

Ahalt, Stanley C.↗

Simultaneous Retrieval of Temperature, Water Vapor and Ozone Atmospheric Profiles from IASI: Compression, De-noising, First Guess Retrieval and Inversion Algorithms

A fast temperature water vapor and ozone atmospheric profile retrieval algorithm is developed for the high spectral resolution Infrared Atmospheric Sounding Interferometer (IASI) space-borne instrument. Compression and de-noising of IASI observations are performed using Principal Component Analysis. This preprocessing methodology also allows, for a fast pattern recognition in a climatological data set to obtain a first guess. Then, a neural network using first guess information is developed to retrieve simultaneously temperature, water vapor and ozone atmospheric profiles. The performance of the resulting fast and accurate inverse model is evaluated with a large diversified data set of radiosondes atmospheres including rare events.

Aires, F.↗

LUNA Condition-Based Monitoring Update: Auto-encoders on Actuator Data [Slides]

An auto-encoder is a neural network trained to output a reconstruction of its input. There is a ‘bottleneck’ layer whose dimension is typically lower than that of the input. Because the input passes through the bottleneck, the auto-encoder must learn to compress (encode) the input into this low-dimensional form, and to decompress (decode) the compressed data into a close approximation of its original form.

97 MATHEMATICS AND COMPUTING↗

Non-linear Post Processing Image Enhancement

A non-linear filter for image post processing based on the feedforward Neural Network topology is presented. This study was undertaken to investigate the usefulness of "smart" filters in image post processing. The filter has shown to be useful in recovering high frequencies, such as those lost during the JPEG compression-decompression process. The filtered images have a higher signal to noise ratio, and a higher perceived image quality. Simulation studies comparing the proposed filter with the optimum mean square non-linear filter, showing examples of the high frequency recovery, and the statistical properties of the filter are given,

Hunt, Shawn↗

Fast convolutional neural networks on FPGAs with hls4ml

We introduce an automated tool for deploying ultra low-latency, low-power deep neural networks with convolutional layers on field-programmable gate arrays (FPGAs). By extending the hls4ml library, we demonstrate an inference latency of 5 µs using convolutional architectures, targeting microsecond latency applications like those at the CERN Large Hadron Collider. Considering benchmark models trained on the Street View House Numbers Dataset, we demonstrate various methods for model compression in order to fit the computational constraints of a typical FPGA device used in trigger and data acquisition systems of particle detectors. In particular, we discuss pruning and quantization-aware training, and demonstrate how resource utilization can be significantly reduced with little to no loss in model accuracy. We show that the FPGA critical resource consumption can be reduced by 97% with zero loss in model accuracy, and by 99% when tolerating a 6% accuracy degradation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Constraining the Multiscale Structure of Geophysical Fields in Machine Learning: The Case of Precipitation

The use of deep-learning algorithms for estimating the value of geophysical variables from remotely sensed information is rapidly expanding. The typical objective function minimized in such algorithms is the mean square error (MSE), which is known to lead to smooth estimates with compressed dynamical range as compared to the true distribution of the variable of interest. Here, we introduce and evaluate alternative objective functions, focusing on the retrieval of precipitation rates from satellite passive microwave radiometric measurements using a deep convolutional neural network. For this testbed application, the results show that explicitly imposing the preservation of the statistical distribution and spatial wavelet power spectrum of the target variable allows to accurately reproduce extreme values and sharp gradients across multiple scales in this study.

58 GEOSCIENCES↗

Woven ceramic matrix composite surrogate model based on physics-informed recurrent neural network

A recurrent neural network (RNN) based surrogate model is developed to emulate the nonlinear constitutive behavior of woven ceramic matrix composites (CMCs) driven by matrix damage at multiple length scales. Physics-informed constraints are introduced into the surrogate model through regularization to ground the prediction in physics and improve its predictive capabilities. Training data is generated using the multiscale generalized method of cells (MSGMC) approach coupled with a matrix damage model. This coupling permits simulating the nonlinear behavior of woven CMCs based on constituent response at the micro-, meso-, and macroscales. The multiscale repeating unit cell is loaded under non-monotonic conditions including multiple load / unload cycles and tension / compression. The fiber volume fraction as well as the intra- and intertow void volume fractions are also varied in the generation of training data. Therefore, the RNN-based surrogate model is tasked with predicting, as a function of variable input strain sequence and fiber and void volume fractions, the resulting stress versus strain response while satisfying physical constraints such as positive semi-definiteness of the tangent stiffness matrix and linear elastic unloading. Further, the trained surrogate model effectively matches the stress versus strain response and successfully predicts the tangent modulus throughout the loading regime. Neural network based surrogate models can offer efficient alternatives to running computationally intensive multiscale material models to simulate the nonlinear response of large structural models. Therefore the presented work provides evidence towards the feasibility of developing, training, and running such models for CMCs with complex architectures, nonlinear multiaxial material response, and under non-monotonic loading conditions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗