Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Convolutional Neural Networks”

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 1,297 records · Page 72
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

Searches for sterile neutrinos with NOvA

Contradictory evidence has been presented on the issue of neutrino mixing between the three known active neutrinos and light sterile neutrino species. Apparent short-baseline neutrino oscillations observed by the LSND and MiniBooNE experiments, the collective evidence of the reactor neutrino anomaly, and the gallium anomaly all point towards sterile neutrinos with mass at the 1 eV level. While these results are tantalizing, they are not conclusive as they are in tension with null results from other short-baseline experiments, and with disappearance searches in longbaseline and atmospheric experiments. The NOvA (NuMI Off-Axis v e Appearance) experiment may help clarify the situation by searching for disappearance of active neutrinos from the NuMI (Neutrinos from the Main Injector) beam over a baseline of 810 km. We describe the method used by NOvA to look for oscillations into sterile neutrinos at the Far Detector (FD) through the disappearance of neutral-current (NC) neutrino events, including preliminary results of this search. In addition, the Near Detector (ND) is well suited for searching for anomalous short-baseline oscillations and probing the LSND and MiniBooNE sterile neutrino allowed regions using a variety of final states. We also present a novel method for selecting samples with high purity at the ND using convolutional neural networks. Furthermore, based on this method, the sensitivity to anomalous short-baseline nt appearance are shown, and searches for anomalous v e appearance and v μ disappearance at the NOvA ND are presented.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

Status of a Deep Learning Based Measurement of the Inclusive Muon Neutrino Charged-current Cross Section in the NOvA Near Detector

NOvA is a long-baseline neutrino oscillation experiment. It uses the NuMI beam from Fermilab and two sampling calorimeter detectors placed off-axis from the beam. The 293 ton Near Detector measures the unoscillated neutrino energy spectrum, which can be used to predict the neutrino energy spectrum observed at the 14 kton Far Detector. The Near Detector also provides an excellent opportunity to measure neutrino interaction cross sections with high statistics, which will benefit current and future long-baseline neutrino oscillation experiments. This analysis implements new algorithms to identify $\nu_{\mu}$ charge-current events by using visual deep learning tools such as convolutional neural networks. We present the status of a measurement of the inclusive $\nu_{\mu}$ CC cross section in the NOvA Near Detector.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

NOvA Short-Baseline Tau Neutrino Appearance Search

Standard three-flavor neutrino oscillations have well explained by a wide range of neutrino experiments. However, the anomalous results, such as electron-antineutrino excess seen by LSND and MiniBooNE do not fit the three-flavor paradigm. This can be explained by an additional fourth flavor sterile neutrino at a larger scale than the existing three flavor neutrinos. The NOvA experiment consists of two finely segmented, liquid scintillator detectors operating 14 .6 mrad off-axis from the NuMI muon-neutrino beam. The Near Detector is located on the Fermilab campus, 1 km from the NuMI target, while the Far Detector is located at Ash River, MN, 810 km from the NuMI target. The NOvA experiment is primarily designed to measure electron-neutrino appearance at the Far Detector using the Near Detector to control systematic uncertainties; however, the Near Detector is well suited for searching for anomalous short-baseline oscillations. This poster will present a novel method for selecting tau neutrino interactions with high purity at the Near Detector using a convolutional neural network. Using this method, the sensitivity to anomalous short-baseline tau-neutrino appearance due to sterile neutrino oscillations will be presented.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2018

Progress Toward a Deep-Learning-Based Search for Low-Energy Electron Neutrinos in MicroBooNE

MicroBooNE is a neutrino experiment based at Fermilab which consists of a liquid argon time-projection chamber in the Booster Neutrino Beam (BNB). The experiment aims to investigate the excess of electron neutrino-like events seen by the MiniBooNE experiment, also located in the BNB. I will describe the status of our deep-learning-based search for low-energy electron neutrino interactions within the MicroBooNE detector and the muon neutrino interactions by which we will constrain them. This analysis features a novel hybrid approach of traditional reconstruction methods combined with the use of convolutional neural networks, a type of deep learning algorithm highly adept at pattern recognition. I will present work on ongoing studies that will characterize and quantify the systematic uncertainties associated with this analysis.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2019

Searching for Short-Range Correlations in $^{40}AR$

The neutrino charged current (CC) cross section is critical for understanding neutrino oscillations as the flavor of a neutrino can only be determined by its CC interaction. As future generations of neutrino detectors will use argon as a target nucleus, it’s important to understand the influence that nuclear effects have on neutrino-nucleon cross sections. Short range correlations (SRCs) are an example of such an effect that are known to contribute non-trivially to the CC cross section. In MicroBooNE, a CC interaction with a correlated nucleon would lead to a 1 muon, 2 proton topology (CC2p). In order to study SRCs in MicroBooNE, It is crucial that the particle identification (PID) techniques used are efficient in identifying CC2p signals from background. Convolutional neural networks (CNN) have shown great promise in their ability to distinguish objects in an image from a noisy background. We present here our work towards the development of a CNN for CC2p event selection and identification.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2019

Classification of Sunyaev-Zel’dovich Galaxy Clusters using Deep Learning

Galaxy clusters are the largest gravitationally bound objects in the universe. One way to find and characterize them is through the Sunyaev-Zel'dovich (SZ) effect in millimeter wave observations of the cosmic microwave background (CMB). From cluster abundance across redshifts and cluster masses, we can infer the growth rate of the universe and thereby extract information about dark energy and cosmic structure evolution. This work represents a first step in applying convolutional neural networks (CNNs) to extract this information. We train a CNN on realistic simulations of the millimeter-wave sky to find galaxy clusters through classification and then to measure their gas mass through regression. We employ relatively simple CNN architectures on a suite of simulated CMB maps, each of which has a different combination of signals and noise: SZ effect, CMB signal, instrument noise, and different foregrounds. In our initial experiments we find varying classification succe ss --- worsening as more of the aforementioned contaminants are included. For regression, the correlation between the predicted and input SZ signal is clear for the case of maps with SZ, CMB, and instrument noise. While the performance does not beat current methods, the CNN is easier to set up and is less susceptible to the influence of highly structured noise terms, such as galactic dust.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

ADVANCED-STEM-BASED DEEP LEARNING FOR SEMANTIC SEGMENTATION OF DEFECTS IN STEELS

In this work, we developed a novel deep convolutional neural network (DCNN) model, called DefectNet, for robust and automated semantic segmentation of three crystallographic defects including line dislocations, precipitates and voids commonly observed in structural metals and alloys. In previous work, we established an experimental protocol for a diffraction contrast imaging scanning transmission electron microscopy (DCI STEM) technique and tailored it specifically for imaging defects in popular iron-based structural alloy. Thus, the DefectNet was trained over a small but high-quality DCI STEM defect image sets of a HT-9 martensitic steel before and after neutron irradiation for 111.8 dpa at 412°C. For the defect quantification that typically takes at least half an hour even for an expert, with a good model the machine learning algorithm can produce more reproducible and reliable quantification in a few seconds. This work demonstrates the feasibility of using deep learning algorithms for fast and accurate defect quantitative analysis.

Machine Learning, Radiation effects, Image analysi↗
DOE OSTI2019

High Performance Geocomputation for Assessing Human Dynamics at Planet Scale

Settlements are key indicators of human presence on the landscape. Large scale mapping of human settlements and their morphology from very high-resolution satellite images is a critical step towards developing an interpretative understanding of population distribution and the sociocultural attributes of the built environment they live in. Convolutional neural network (CNN) based Deep Learning experiments indicates that such computations can be scaled to some of the largest high-performance computing (HPC) architectures. While early results are encouraging for developing settlement and corresponding population maps at unprecedented speed and spatial resolutions, characterizing human dynamics at planet scale with high temporal resolution will require the community to develop novel geocomputational infrastructures and ecosystems.

Bhaduri, Budhu↗
DOE OSTI2019

In-Place Zero-Space Memory Protection for CNN

Convolutional Neural Networks (CNN) are being actively explored for safetycritical applications such as autonomous vehicles and aerospace, where it is essential to ensure the reliability of inference results in the presence of possible memory faults. Traditional methods such as error correction codes (ECC) and Triple Modular Redundancy (TMR) are CNN-oblivious and incur substantial memory overhead and energy cost. This paper introduces in-place zero-space ECC assisted with a new training scheme weight distribution-oriented training. The new method provides the first known zero space cost memory protection for CNNs without compromising the reliability offered by traditional ECC.

Guan, Hui↗
DOE OSTI2019

GMLS-Nets: Scientific Machine Learning Methods for Unstructured Data

Data fields sampled on irregularly spaced points arise in many applications in the sciences and engineering. For regular grids, Convolutional Neural Networks (CNNs) have been successfully used to gaining benefits from weight sharing and invariances. We generalize CNNs by introducing methods for data on unstructured point clouds based on Generalized Moving Least Squares (GMLS). GMLS is a nonparametric technique for estimating linear bounded functionals from scattered data, and has recently been used in the literature for solving partial differential equations. By parameterizing the GMLS estimator, we obtain learning methods for operators with unstructured stencils. In GMLS-Nets the necessary calculations are local, readily parallelizable, and the estimator is supported by a rigorous approximation theory. We show how the framework may be used for unstructured physical data sets to perform functional regression to identify associated differential operators and to regress quantities of interest. The results suggest the architectures to be an attractive foundation for data-driven model development in scientific machine learning applications

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

GMLS-Nets: A Framework for Learning from Unstructured Data

Data fields sampled on irregularly spaced points arise in many applications in the sciences and engineering. For regular grids, Convolutional Neural Networks (CNNs) have been successfully used to gaining benefits from weight sharing and invariances. We generalize CNNs by introducing methods for data on unstructured point clouds based on Generalized Moving Least Squares (GMLS). GMLS is a non-parametric technique for estimating linear bounded functionals from scattered data, and has recently been used in the literature for solving partial differential equations. By parameterizing the GMLS estimator, we obtain learning methods for operators with unstructured stencils. In GMLS-Nets the necessary calculations are local, readily parallelizable, and the estimator is supported by a rigorous approximation theory. We show how the framework may be used for unstructured physical data sets to perform functional regression to identify associated differential operators and to regress quantities of interest. The results suggest the architectures to be an attractive foundation for data-driven model development in scientific machine learning applications

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Artificial Intelligence and Computer Vision for Cost-Effective Benthic Habitat Characterizations

Machine learning applied to computer vision and pattern recognition is a type of artificial intelligence that has advanced rapidly in the last 10 to 15 years, spurred forward by breakthroughs in deep convolutional neural networks. These state-of-the-art methods are poised to become widely used in environmental monitoring applications as a result of the increasing abundance of data available from different imaging platforms (e.g., fixed-point cameras, drone surveys, high-resolution satellite data) that can be analyzed to observe, model, and understand environmental conditions. Computer vision and pattern recognition (CVPR) tools advance our ability to use imagery and camera-based tools in cost-effective ways for environmental monitoring. Although these techniques offer great potential, some challenges remain, such as the need for large sets of labeled images for model training and validation and optimized hardware and software to ensure that the models can be trained effectively and in a reasonable amount of time. Here, we have overcome many of these rate-limiting challenges by using a diverse image library built across multiple projects coupled with staff expertise and onsite computing resources. We are working toward a fully automated SPI processing system and also are moving to develop CVPR analytical tools for other imaging platforms and data sets.

02 PETROLEUM↗
DOE OSTI2020

Deep Learning for Automated Identification of Eels in Sonar Data

Freshwater eels, such as the American eel (Anguilla rostrata) present numerous challenges related to safe downstream fish passage at hydroelectric facilities. One of those challenges is effective monitoring of their abundance, movements, and behavior to facilitate design and operation of eel protection and passage facilities. A previous EPRI study documented the ability of human analysts to reliably identify American eels in data obtained with a 1100/1800 kHz, multibeam sonar. This report describes a project to develop deep learning (a subset of artificial intelligence) tools to automate the time-consuming, subjective process of eel identification in multibeam sonar data. The project exploited new data collected in the laboratory and the existing data from the prior EPRI field study to develop and test deep learning and other data analytic tools, including wavelet filtering, differencing for static object removal, and convolutional neural network analysis. The analysis of the laboratory data demonstrated feasibility of the approach, revealed object characteristics observed with the sonar that distinguish eels from similarly sized and shaped acoustic targets, and provided additional data for algorithm selection and training. Deep learning algorithms trained and tested on the laboratory data alone achieved accuracy rates of greater than 98% when classifying acoustic images of eels and similar-sized neutrally buoyant sticks. The algorithm trained and tested on the pre-existing field data alone, and yielded classification accuracy of 9.3% false positives and 13.3% false negatives when distinguishing between eels and sticks/PVC pipes based on video clips (i.e., multiple, consecutive images). This performance is comparable to the classification accuracy achieved by human analysts in the prior study. The deep learning algorithm trained on a combination of video clips obtained in the laboratory and the field and tested on video clips from the field, was able to distinguish eels from sticks and PVC pipes (a river debris analog) of similar size with 100% accuracy. Outreach to the hardware, software, and end-user communities early in the project helped to identify needs and specify the application space. Outreach to those communities at the end of the project communicated project results and opportunities for further development. The project achieved proof of concept for automated identification of eel in multibeam sonar data. Future work should focus on acquisition of additional data for more robust algorithm training and testing; modification of the software tools to accommodate multiple acoustic targets in the acoustic field at a given time; identification of additional object classes; incorporation of motion in the object identification and classification algorithms; operationalizing the software tools, including integration with other existing sonar data analysis tools; and partnering with hardware and software providers for distribution of the software tools with their commercial products.

13 HYDRO ENERGY↗
DOE OSTI2019

A Search for Anomalous Muon Neutrino Disappearance and Tau Neutrino Appearance using NOvA Near Detector

The NOvA experiment consists of two finely segmented, liquid scintillator detectors operating 14.6 mrad off-axis from the NuMI muon-neutrino beam. The Near Detector (ND) is located in the Fermilab campus, 1 km from the NuMI target, while the Far Detector (FD) is located at Ash River, MN, 810 km from the NuMI target. The NOvA experiment is primarily designed to measure electron-neutrino appearance at the FD using the ND to control systematic uncertainties. However, the ND is well suited to search for anomalous short-baseline (SBL) oscillations. The standard three-flavor neutrino oscillations have well explained by a wide range of neutrino experiments. The anomalous results, such as electron-antineutrino excess seen by LSND and MiniBooNE do not fit the three-flavor paradigm. This can be explained by an additional fourth flavor sterile neutrino at a larger mass scale than the existing three flavor neutrinos. This thesis presents a novel method for selecting tau neutrino interact ions with high purity at the ND using a convolutional neural network (CNN). Using this method, the preliminary sensitivities of anomalous SBL muon-neutrino disappearance and tau-neutrino appearance due to sterile neutrino oscillations in NOvA are shown.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Analysis of MINERvA tunings of GENIE based on Hadron Multiplicity using Machine Learning

Scientists with the MINERvA neutrino interaction experiment have long studied distributions of hadrons (protons and pions) to improve the models for these reactions. Until now, most work has relied on easy to find, energetic, and well tracked hadrons. But in order to take full advantage of MINERνA data it is necessary to develop an efficient enough tool to recognize hadron tracks with lower energy, more transverse angle and more challenging patterns of activity. In this analysis, a Machine Learning tool based on Convolutional Neural Network and Transfer Learning has been developed to tackle that issue. It is found to be extremely capable of recognizing hadron activities especially when there are few and not very energetic hadrons, the accuracy is 76%. But it is also found that this network is very reluctant to classify too many hadrons. The classification is also affected by the presence of other particles in the event such photons, hadrons below a threshold or neutrons. This network is used to compare hadron multiplicity of data to the MC models and distinguish between different MC tunes. Some MINERνA tunes are explored in this analysis, but none of them described the data very well, especially not for all different hadrons multiplicities at once. Four combinations of those tunes were more successful, but none of them could be recognized as the best one. Nevertheless, this analysis develops an ML network that sets the basis of further exploration of hadron multiplicity as a tool to analyze MC tunes at MINERνA and beyond.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Measurement of Electron Neutrino and Antineutrino Appearance with the NO{$\nu$}A Experiment

As a long-baseline neutrino oscillation experiment, the NuMI Off-axis $\nu_e$ Appearance (NOvA) experiment aims at studying neutrino physics by measuring neutrino oscillation parameters using the neutrino flux from the Main Injector (NuMI) beam. It has two functionally identical detectors. The near detector is onsite at Fermi National Accelerator Laboratory. The far detector is 810 km away from the source of neutrinos and antineutrinos, at Ash River, Minnesota. At the near detector, muon neutrinos or antineutrinos, before significant oscillations take place, are used to correct the Monte Carlo simulation. At the far detector, the neutrino and antineutrino fluxes after significant oscillations have happened are measured and analyzed to study neutrino oscillation. The NOvA experiment is sensitive to the values of $\sin^2{\theta_{23}}$, $\Delta m^2_{32}$, and $\delta_{CP}$. The latest values from the NOvA 2020 analysis are as follows: $\sin^2{\theta_{23}}= 0.57^{+0.03}_{-0.04}$, $\Delta m^2_{32} = (2.41 \pm 0.07)\times 10^{-3}$\,eV$^2/c^4$, and $\delta_{CP}=0.82\pi$ with a wide 1$\sigma$ interval of uncertainty. My study is focused on the neutrino oscillation analysis with NOvA, including detector light model tuning, particle classification with convolutional neural network, electron neutrino and antineutrino energy reconstruction, and oscillation background estimation. Most of my studies have been used in the latest NOvA publication and the NOvA 2020 analysis.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Multimodal Data Representation with Deep Learning for Extracting Cancer Characteristics from Clinical Text

This paper presents a multimodal data representation to improve the performance of deep learning models for extracting cancer key characteristics from unstructured text in pathology reports. Specifically, in addition to using the text as the input to deep learning models, we use concept unique identifiers (CUIs) as another source of information to the models. We analyze the performance of different text and CUI data representations, including word embeddings and bag of embeddings (BOE), with a convolutional neural network (CNN) and a fully connected multilayer perceptron neural network (MLP-NN). The high level document embeddings from text and CUI inputs are combined by concatenating them and then applying a classifier. The model is used for extracting cancer subsite and histology from pathology reports. These two classification tasks have a large number of labels, i.e. 317 for subsite and 556 for histology, with extreme class imbalance. We compare the performance of the developed DL models across the two tasks based on micro- and macro-F1 scores. The evaluation shows that a multi-channel DL model that utilizes text represented by word embeddings and CUIs represented by BOE outperforms other DL models. Also, this approach significantly improves the model performance on low prevalence classes.

Alawad, Mohammed↗
DOE OSTI2019

Toward A High-Performance Emulation Platform for Brain-Inspired Intelligent Systems Exploring Dataflow-Based Execution Model and Beyond

Brain-inspired computing is a novel computing technology based on neural morphological engineering, which draws lessons from methods of human brain information processing and storage. Combining with the high-performance computing (HPC) platform, they constitute the foundation of general artificial intelligence. However, current brain HPC platforms generally suffer from slow speed, poor scalability, and high energy consumption, which severely restrain its potential and circumscribe the development of general artificial intelligence. The dataflow model was first proposed in the 1970s, providing a novel idea for the development of HPC. In addition, the dataflow model shares similar information processing mechanisms with human's neural system, which makes dataflow models naturally suit the emulation of brain-inspired computing. Based on the contemporary progress of the dataflow model, the Codelet model was proposed. Through a fine-grained asynchronous program execution and resource allocation, the Codelet model successfully realized the distributed computing on the heterogeneous system, effectively improved the computing power and speed, and open up a new path to overcome the shortcomings of the existing high-performance computing technology. We propose a dataflow-based emulation platform, aiming at providing high-performance computing technology support for general brain-inspired intelligent system, as well as using characteristics of dataflow models to fully explore the potential of brain-inspired intelligence. As an example, we will select a convolutional neural network (LeNet5) that already has a spectacular user base to initially verify the superiority and feasibility of our proposal.

brain-inspired intelligent system↗