Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “explainable 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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Enhancing Neural Network Decision-Making with Variational Autoencoders

Machine intelligence has been used to tackle increasingly complex problems and deep learning solutions are at the forefront of tackling these problems. In general, these architectures have a great number of parameters that are methodically updated in training. The vast number and complexity of deep neural networks makes it very difficult to decipher the inner workings of the neurons and layers that make up the network. This paper posits that trustworthiness and trust in autonomous systems are increased through eXplainable Artificial Intelligence (XAI) and presents a method that enhances the explainability and understanding of a neural network decision. We leverage variational autoencoders to produce human interpretable features from complex data sets. We show that the explainable features can then be used for machine learning applications. This explainability encourages people to be more inclined to justifiably trust machine decision-making.

Loc Tran↗
DOE OSTI2023

A detailed study of interpretability of deep neural network based top taggers

Abstract Recent developments in the methods of explainable artificial intelligence (XAI) allow researchers to explore the inner workings of deep neural networks (DNNs), revealing crucial information about input–output relationships and realizing how data connects with machine learning models. In this paper we explore interpretability of DNN models designed to identify jets coming from top quark decay in high energy proton–proton collisions at the Large Hadron Collider. We review a subset of existing top tagger models and explore different quantitative methods to identify which features play the most important roles in identifying the top jets. We also investigate how and why feature importance varies across different XAI metrics, how correlations among features impact their explainability, and how latent space representations encode information as well as correlate with physically meaningful quantities. Our studies uncover some major pitfalls of existing XAI methods and illustrate how they can be overcome to obtain consistent and meaningful interpretation of these models. We additionally illustrate the activity of hidden layers as neural activation pattern diagrams and demonstrate how they can be used to understand how DNNs relay information across the layers and how this understanding can help to make such models significantly simpler by allowing effective model reoptimization and hyperparameter tuning. These studies not only facilitate a methodological approach to interpreting models but also unveil new insights about what these models learn. Incorporating these observations into augmented model design, we propose the particle flow interaction network model and demonstrate how interpretability-inspired model augmentation can improve top tagging performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Quantifying the generalization error in deep learning in terms of data distribution and neural network smoothness

We report the accuracy of deep learning, i.e., deep neural networks, can be characterized by dividing the total error into three main types: approximation error, optimization error, and generalization error. Whereas there are some satisfactory answers to the problems of approximation and optimization, much less is known about the theory of generalization. Most existing theoretical works for generalization fail to explain the performance of neural networks in practice. To derive a meaningful bound, we study the generalization error of neural networks for classification problems in terms of data distribution and neural network smoothness. We introduce the cover complexity (CC) to measure the difficulty of learning a data set and the inverse of the modulus of continuity to quantify neural network smoothness. A quantitative bound for expected accuracy/error is derived by considering both the CC and neural network smoothness. Although most of the analysis is general and not specific to neural networks, we validate our theoretical assumptions and results numerically for neural networks by several data sets of images. The numerical results confirm that the expected error of trained networks scaled with the square root of the number of classes has a linear relationship with respect to the CC. We also observe a clear consistency between test loss and neural network smoothness during the training process. In addition, we demonstrate empirically that the neural network smoothness decreases when the network size increases whereas the smoothness is insensitive to training dataset size.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2019

Testing a Neural Network Accelerator on a High-Altitude Balloon

The cognitive communications project has been working to re d machine learning approaches to support their deployment and sustained use in space environments. It has historically been difficult to implement such techniques on space platforms, however, due to the computational requirements they levy onto general-purpose avionics hardware. While technologies exist to accelerate the computation of aspects of neural networks, such platforms have not historically been deployed in space environments. Given that testing payloads in such environments can be both cost- and time-prohibitive, high-altitude balloons can be used as a way to approximate a space environment at a much lower cost, thus providing a cost-effective way in which to test newer approaches to hardware acceleration for artificial intelligence which may be deployed onto spacecraft more directly. This paper describes a successful test of a commercial off- the-shelf neural network accelerator on a high-altitude balloon. It begins by explaining our selection criteria when evaluating different commercial neural network acceleration techniques: primary considerations include size, weight, and power (SWaP) as well as ease of integration. Next, the paper describes the development and implementation of an experimental flight test platform: flight and ground components are discussed. Afterward, the paper discusses the experimental payload itself: this includes the experimental procedure as well as the specific image and method used for testing. Finally, the paper concludes with an evaluation of both the experimental device tested at altitude as well as the flight test framework itself, identifying how the existing platform can be used to continue tes g commercial off-the-shelf (COTS) solutions for acceleration.

Clark, Gilbert↗
NASA NTRS1993

On neural networks in identification and control of dynamic systems

This paper presents a discussion of the applicability of neural networks in the identification and control of dynamic systems. Emphasis is placed on the understanding of how the neural networks handle linear systems and how the new approach is related to conventional system identification and control methods. Extensions of the approach to nonlinear systems are then made. The paper explains the fundamental concepts of neural networks in their simplest terms. Among the topics discussed are feed forward and recurrent networks in relation to the standard state-space and observer models, linear and nonlinear auto-regressive models, linear, predictors, one-step ahead control, and model reference adaptive control for linear and nonlinear systems. Numerical examples are presented to illustrate the application of these important concepts.

Phan, Minh↗
DOE OSTI2021

Leveraging Community and Author Context to Explain the Performance and Bias of Text-Based Deception Detection Models

Deceptive news posts shared in online communities can be detected with NLP models, and much recent research has focused on the development of such models. In this work, we use characteristics of online communities and authors --- the context of how and where content is posted --- to explain the performance of a neural network deception detection model and identify sub-populations who are disproportionately affected by model accuracy or failure. We examine who is posting the content, and where the content is posted to. We find that while author characteristics are better predictors of deceptive content than community characteristics, both characteristics are strongly correlated with model performance. Traditional performance metrics such as F1 score may fail to capture poor model performance on isolated sub-populations such as specific authors, and as such, more nuanced evaluation of deception detection models is critical.

machine learning (ML), machine learning explanatio↗
DOE OSTI2023

Deep material network via a quilting strategy: visualization for explainability and recursive training for improved accuracy

Recent developments integrating micromechanics and neural networks offer promising paths for rapid predictions of the response of heterogeneous materials with similar accuracy as direct numerical simulations. The deep material network is one such approaches, featuring a multi-layer network and micromechanics building blocks trained on anisotropic linear elastic properties. Once trained, the network acts as a reduced-order model, which can extrapolate the material’s behavior to more general constitutive laws, including nonlinear behaviors, without the need to be retrained. However, current training methods initialize network parameters randomly, incurring inevitable training and calibration errors. Here, we introduce a way to visualize the network parameters as an analogous unit cell and use this visualization to “quilt” patches of shallower networks to initialize deeper networks for a recursive training strategy. The result is an improvement in the accuracy and calibration performance of the network and an intuitive visual representation of the network for better explainability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Real-time biomass feedstock particle quality detection using image analysis and machine vision

Abstract A common and costly challenge in the nascent biorefinery industry is the consistent handling and conveyance of biomass feedstock materials, which can vary widely in their chemical, physical, and mechanical properties. Solutions to cope with varying feedstock qualities will be required, including advanced process controls to adjust equipment and reject feedstocks that do not meet a quality standard. In this work, we present and evaluate methods to autonomously assess corn stover feedstock quality in real time and provide data to process controls with low-cost camera hardware. We explore the use of neural networks to classify feedstocks based on actual processing behavior and pixel matrix feature parameterization to further assess particle attributes that may explain the variable processing behavior. We used the pretrained ResNet neural network coupled with a gated recurrent unit (GRU) time-series classifier trained on our image data, resulting in binary classification of feedstock anomalies with favorable performance. The textural aspects of the image data were statistically analyzed to determine if the textural features were predictive of operational disruptions. The significant textural features were angular second moment, prominence, mean height of surface profile, mean resultant vector, shade, skewness, variation of the polar facet orientation, and direction of azimuthal facets. Expansion of these models is recommended across a wider variety of labeled feedstock images of different qualities and species to develop a more robust tool that may be deployed using low-cost cameras within biorefineries.

09 BIOMASS FUELS↗
NASA NTRSDate not supplied

Case Study: Analysis of Autonomous Center line Tracking Neural Networks

Deep neural networks have gained widespread usage in a number of applications. However, limitations such as lack of explainability and robustness inhibit building trust in their behavior, which is crucial in safety critical applications such as autonomous driving. Therefore, techniques which aid in understanding and providing guarantees for neural network behavior are the need of the hour. In this paper, we present a case study applying a recently proposed technique, Prophecy, to analyze the behavior of a neural network model, provided by our industry partner and used for autonomous guiding of airplanes on taxi runways. This regression model takes as input an image of the runway and produces two outputs, cross-track error and heading error, which represent the position of the plane relative to the center line. We use the Prophecy tool to extract neuron activation patterns for the correctness and safety properties of the model. We show the use of these patterns to identify features of the input that explain correct and incorrect behavior. We also use the patterns to provide guarantees of consistent behavior. We explore a novel idea of using sequences of images (instead of single images) to obtain good explanations and identify regions of consistent behavior.

Deep Neural Networks↗
DOE OSTI2022

Explainable machine learning in materials science

Abstract Machine learning models are increasingly used in materials studies because of their exceptional accuracy. However, the most accurate machine learning models are usually difficult to explain. Remedies to this problem lie in explainable artificial intelligence (XAI), an emerging research field that addresses the explainability of complicated machine learning models like deep neural networks (DNNs). This article attempts to provide an entry point to XAI for materials scientists. Concepts are defined to clarify what explain means in the context of materials science. Example works are reviewed to show how XAI helps materials science research. Challenges and opportunities are also discussed.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Quantum Neural Networks: Issues, Training, and Applications

Our work in the field aims at explaining the limitations and expressive power of Quantum Machine Learning models, as well as finding feasible training algorithms that could be implemented in near-term Quantum Computers. The promise of Quantum Machine Learning is that by incorporating quantum effects, such as entanglement, into machine learning models researchers can improve model performance and understand more complex datasets. This pledge is particularly pronounced in the design of Quantum neural networks (QNNs), a promising framework for creating quantum algorithms, that promise to outperform classical models by combining the speedups of quantum computation with the widespread successes of deep learning. We show that applying this approach alone to quantum deep learning is problematic given that an excess of entanglement between the hidden and visible layers can destroy the predictive power of our QNN models. We address the barren plateau problem by suggesting the use of a generative, unbounded, nonlinear loss function with simple gradients. The loss function quantifies how much the quantum states generated by the QNNs differ from the data and the goal during training is to minimize it. Finally, we showcase how to use generative training to construct a "classical-quantum" neural network to accurately interpolate between the ground states of a Molecular Hamiltonian, a central question in Quantum Chemistry.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Unsupervised Azimuth Estimation of Solar Arrays in Low-Resolution Satellite Imagery through Semantic Segmentation and Hough Transform

This paper explains the use of a convolutional neural network (CNN) to segment solar panels in a satellite image containing solar arrays, and extract associated metadata from the arrays. A novel unsupervised technique is introduced to estimate the azimuth of each individual solar panel from the predicted mask of the convolutional neural network. This pipeline was developed with the aim of extracting necessary metadata for a solar installation, using only a set of latitude–longitude coordinates. Azimuth prediction results for 669 individual solar installations associated with 387 sites located across the United States are provided. A mean average error and median average error of 21.65 degrees and 1.0 degrees were obtained, respectively, when predicting the azimuth of the solar fleet data set, with about 80% of the results within an error of zero degrees of the ground truth azimuth value and about 85% within an error of 25 degrees. The predicted azimuth was then used to estimate the energy conversion of the solar arrays. Results show a 90.9 and 90.6 R-squared value for estimating alternating current (AC) and direct current (DC) energy, respectively, and a mean absolute percentage error (MAPE) of 1.70% in estimating the alternating current (AC) energy using the fully automated algorithm.

14 SOLAR ENERGY↗
DOE OSTI2022

A Review on the State of the Art of Machine Learning and Satellite Imaging: Detecting Scene Changes in Selected Nuclear Fuel Cycle Datasets

The timely detection of clandestine nuclear facilities is one of the greatest challenges faced by the International Atomic Energy Agency’s (IAEA). Idaho National Laboratory is currently applying machine learning (ML) to existing satellite imagery (SI) datasets to find facilities within the nuclear fuel life cycle, with primary focus placed on identifying critical predecessor (i.e., fuel fabrication and fuel enrichment) and successor (i.e., nuclear power plants) facilities. This could provide a satellite image methodology that the IAEA could leverage to discover clandestine facilities. The work presented in this paper describes the evolution of a workflow developed by this team for object detection related to critical infrastructure by expanding that workflow for the purpose of identifying nuclear fuel cycle components and automating dependency assessments. This will be done using two methods housed within a single pipeline. The first method involves implementing a DenseNet161 convolutional neural network to classify the images and explain the results using Local Interpretable Model-Agnostic Explanations (LIME). The second method implements You Only Look Once version 5 (YoloV5), to detect objects within images, provide a probability for the detection, and provide a bounding box that corresponds to the object of interest. The results of this work are anticipated to provide a clear picture of this portion of the nuclear fuel cycle and perform as a stand-alone tool for image assessment that can be expanded to additional fuel cycle components and implemented in international safeguards and national security domains. This capability addresses the IAEA’s need to detect undeclared nuclear materials and activities within a state while encompassing the entire nuclear fuel cycle.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

An Explainable Machine-Learning Model for Compensatory Reserve Measurement: Methods for Feature Selection and the Effects of Subject Variability

Tracking vital signs accurately is critical for triaging a patient and ensuring timely therapeutic intervention. The patient’s status is often clouded by compensatory mechanisms that can mask injury severity. The compensatory reserve measurement (CRM) is a triaging tool derived from an arterial waveform that has been shown to allow for earlier detection of hemorrhagic shock. However, the deep-learning artificial neural networks developed for its estimation do not explain how specific arterial waveform elements lead to predicting CRM due to the large number of parameters needed to tune these models. Alternatively, we investigate how classical machine-learning models driven by specific features extracted from the arterial waveform can be used to estimate CRM. More than 50 features were extracted from human arterial blood pressure data sets collected during simulated hypovolemic shock resulting from exposure to progressive levels of lower body negative pressure. A bagged decision tree design using the ten most significant features was selected as optimal for CRM estimation. This resulted in an average root mean squared error in all test data of 0.171, similar to the error for a deep-learning CRM algorithm at 0.159. By separating the dataset into sub-groups based on the severity of simulated hypovolemic shock withstood, large subject variability was observed, and the key features identified for these sub-groups differed. This methodology could allow for the identification of unique features and machine-learning models to differentiate individuals with good compensatory mechanisms against hypovolemia from those that might be poor compensators, leading to improved triage of trauma patients and ultimately enhancing military and emergency medicine.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Results From Invoking Artificial Neural Networks to Measure Insider Threat Detection & Mitigation

Advances on differentiating between malicious intent and natural “organizational evolution” to explain observed anomalies in operational workplace patterns suggest benefit from evaluating collective behaviors observed in the facilities to improve insider threat detection and mitigation (ITDM). Advances in artificial neural networks (ANN) provide more robust pathways for capturing, analyzing, and collating disparate data signals into quantitative descriptions of operational workplace patterns. In response, a joint study by Sandia National Laboratories and the University of Texas at Austin explored the effectiveness of commercial artificial neural network (ANN) software to improve ITDM. Overall, this research demonstrates the benefit of learning patterns of organizational behaviors, detecting off-normal (or anomalous) deviations from these patterns, and alerting when certain types, frequencies, or quantities of deviations emerge for improving ITDM. Evaluating nearly 33,000 access control data points and over 1,600 intrusion sensor data points collected over a nearly twelve-month period, this study's results demonstrated the ANN could recognize operational patterns at the Nuclear Engineering Teaching Laboratory (NETL) and detect off-normal behaviors—suggesting that ANNs can be used to support a data-analytic approach to ITDM. Several representative experiments were conducted to further evaluate these conclusions, with the resultant insights supporting collective behavior-based analytical approaches to quantitatively describe insider threat detection and mitigation.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗