SEARCH · Engineering Papers
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.
We report the results of the first search for the decay B$^{0}_{s}$ → π 0 π 0 using 121.4 fb -1 of data collected at the Y(5S) resonance with the Belle detector at the KEKB asymmetric-energy e + e - collider. We observe no signal and set a 90% confidence level upper limit of 7.7 ×10 -6 on the B$^{0}_{s}$ → π 0 π 0 decay branching fraction.
Predicting vehicle mass is critical to accurately estimate energy use and emissions of commercial trucks. However, data from vehicle telematics is often not at sufficient temporal resolution or accuracy for use in model-based detection methods. In this work, a new statistical mass prediction technique is described for heavy-duty vehicles that incorporates the use Gaussian Belief Propagation (GBP) for probabilistic inference. Similar to Bayesian inference models, the GBP model typically requires less labeled training data than other contemporary machine learning techniques. First, a factor graph is constructed, and a set of Gaussian belief nodes with associated means and variances are fitted to the training data. To better handle noisy input data, the GBP mass prediction model utilizes a k-nearest factors (kNF) algorithm for probabilistic inference on unseen testing data. The proposed method is compared with a classical weighted k-nearest neighbors (kNN) regressor. This statistical kNF-GBP model works even with low-quantity, low-quality initial training data, while being capable of realtime mass estimation. Unlike the kNN regressor, the GBP model produces a measure of uncertainty with its predictions. The proposed method is validated using curve-sampled driving data collected from multiple cloud-connected Class 8 regional haul diesel trucks. Both the kNN regressor and the kNF-GBP mass prediction model were able to predict payload mass with coefficients of determination above 0.97 with minimal data preprocessing.
This package implements Bayesian Neural Network (BNN) based surrogate models for multi-modal data. These models are fit using a custom Variational Bayes procedure leveraging conjugate posteriors in the last layer of the network, offering more accurate predictions with better-calibrated uncertainty quantification than mean-field Variational Bayes.
Abstract Background The preservation of soil health is a critical challenge in the 21st century due to its significant impact on agriculture, human health, and biodiversity. We provide one of the first comprehensive investigations into the predictive potential of machine learning models for understanding the connections between soil and biological phenotypes. We investigate an integrative framework performing accurate machine learning-based prediction of plant performance from biological, chemical, and physical properties of the soil via two models: random forest and Bayesian neural network. Results Prediction improves when we add environmental features, such as soil properties and microbial density, along with microbiome data. Different preprocessing strategies show that human decisions significantly impact predictive performance. We show that the naive total sum scaling normalization that is commonly used in microbiome research is one of the optimal strategies to maximize predictive power. Also, we find that accurately defined labels are more important than normalization, taxonomic level, or model characteristics. ML performance is limited when humans can’t classify samples accurately. Lastly, we provide domain scientists via a full model selection decision tree to identify the human choices that optimize model prediction power. Conclusions Our study highlights the importance of incorporating diverse environmental features and careful data preprocessing in enhancing the predictive power of machine learning models for soil and biological phenotype connections. This approach can significantly contribute to advancing agricultural practices and soil health management.
We present iHOMER, an iterative version of the HOMER method to extract Lund fragmentation functions from experimental data. Through iterations, we address the information gap between latent and observable phase spaces and systematically remove bias. To quantify uncertainties on the inferred weights, we use a combination of Bayesian neural networks and uncertainty-aware regression. We find that the combination of iterations and uncertainty quantification produces well-calibrated weights that accurately reproduce the data distribution. A parametric closure test shows that the iteratively learned fragmentation function is compatible with the true fragmentation function.
Deep learning is used in many applications in the physical sciences. In those sciences we are used to having uncertainty on every measurement or prediction. In the last years, many methods have been put forth for uncertainty quantification: Bayesian Neural Networks, Concrete Dropout, and Deep Ensembles. Uncertainty in deep learning is often divided into aleatoric or irreducible and epistemic or reducible. The problem remains: Which uncertainty quantification method should be chosen? How are those results interpreted?
Probabilistic and Bayesian neural networks have long been proposed as a method to incorporate uncertainty about the world (both in training data and operation) into artificial intelligence applications. One approach to making a neural network probabilistic is to leverage a Monte Carlo sampling approach that samples a trained network while incorporating noise. Such sampling approaches for neural networks have not been extensively studied due to the prohibitive requirement of many computationally expensive samples. While the development of future microelectronics platforms that make this sampling more efficient is an attractive option, it has not been immediately clear how to sample a neural network and what the quality of random number generation should be. This research aimed to start addressing these two fundamental questions by examining basic “off the shelf” neural networks can be sampled through a few different mechanisms (including synapse “dropout” and neuron “dropout”) and examine how these sampling approaches can be evaluated both in terms of evaluating algorithm effectiveness and the required quality of random numbers.
This project evaluated the use of emerging spintronic memory devices for robust and efficient variational inference schemes. Variational inference (VI) schemes, which constrain the distribution for each weight to be a Gaussian distribution with a mean and standard deviation, are a tractable method for calculating posterior distributions of weights in a Bayesian neural network such that this neural network can also be trained using the powerful backpropagation algorithm. Our project focuses on domain-wall magnetic tunnel junctions (DW-MTJs), a powerful multi-functional spintronic synapse design that can achieve low power switching while also opening the pathway towards repeatable, analog operation using fabricated notches. Our initial efforts to employ DW-MTJs as an all-in-one stochastic synapse with both a mean and standard deviation didn’t end up meeting the quality metrics for hardware-friendly VI. In the future, new device stacks and methods for expressive anisotropy modification may make this idea still possible. However, as a fall back that immediately satisfies our requirements, we invented and detailed how the combination of a DW-MTJ synapse encoding the mean and a probabilistic Bayes-MTJ device, programmed via a ferroelectric or ionically modifiable layer, can robustly and expressively implement VI. This design includes a physics-informed small circuit model, that was scaled up to perform and demonstrate rigorous uncertainty quantification applications, up to and including small convolutional networks on a grayscale image classification task, and larger (Residual) networks implementing multi-channel image classification. Lastly, as these results and ideas all depend upon the idea of an inference application where weights (spintronic memory states) remain non-volatile, the retention of these synapses for the notched case was further interrogated. These investigations revealed and emphasized the importance of both notch geometry and anisotropy modification in order to further enhance the endurance of written spintronic states. In the near future, these results will be mapped to effective predictions for room temperature and elevated operation DW-MTJ memory retention, and experimentally verified when devices become available.
Current and future ground-based cosmological surveys, such as the Dark Energy Survey (DES), and the Vera Rubin Observatory Legacy Survey of Space and Time (LSST), are predicted to discover thousands to tens of thousands of strong gravitational lenses. The large number of strong lenses discoverable in future surveys will make strong lensing a highly competitive and complementary cosmic probe. However, conventional lens modeling techniques are unable to scale up to the sheer number of lenses that will be discovered through upcoming surveys. Therefore, the use of automated lens analysis techniques is necessary. We demonstrate that machine learning methods can be used to automate the inference of informative model posteriors of strong lensing systems in ground-based surveys with credible uncertainty estimation. We present two Simulation-Based Inference (SBI) approaches for lens parameter estimation of galaxy-galaxy lenses. We demonstrate applications of Neural Posteriors Estima tors (NPEs) and Bayesian Neural Network (BNNs) to automate the inference of a 12-parameter lensing system for DES-like ground-based imaging data. We apply a suite of diagnostics (e.g., posterior coverage and SBC) to validate the performance of our methods. We find that NPEs outperform the BNN, producing posterior distributions that are for the most part both more accurate and more precise; in particular, several source-light model parameters are systematically biased in the BNN implementation.
This is ongoing work towards developing and integrating a Bayesian Neural Network (BNN) model for Particle-in-Cell (PIC) stress in MFiX legacy code.
Explore the source record for details and available documents.
A main challenge for integrating the intermittent photovoltaic (PV) power generation remains the accuracy of day-ahead forecasts and the establishment of robust performing methods. The purpose of this work is to address these technological challenges by evaluating the day-ahead PV production forecasting performance of different machine learning models under different supervised learning regimes and minimal input features. Specifically, the day-ahead forecasting capability of Bayesian neural network (BNN), support vector regression (SVR), and regression tree (RT) models was investigated by employing the same dataset for training and performance verification, thus enabling a valid comparison. The training regime analysis demonstrated that the performance of the investigated models was strongly dependent on the timeframe of the train set, training data sequence, and application of irradiance condition filters. Furthermore, accurate results were obtained utilizing only the measured power output and other calculated parameters for training. Consequently, useful information is provided for establishing a robust day-ahead forecasting methodology that utilizes calculated input parameters and an optimal supervised learning approach. Finally, the obtained results demonstrated that the optimally constructed BNN outperformed all other machine learning models achieving forecasting accuracies lower than 5%.
We introduce a hierarchical framework that combines ab initio many-body calculations with a Bayesian neural network, developing emulators capable of accurately predicting nuclear properties across isotopic chains simultaneously and being applicable to different regions of the nuclear chart. We benchmark our developments using the oxygen isotopic chain, achieving accurate results for ground-state energies and nuclear charge radii, while providing robust uncertainty quantification. Our framework enables global sensitivity analysis of nuclear binding energies and charge radii with respect to the low-energy constants that describe the nuclear force.
We propose a distributed agent architecture (DAA) that can support a variety of paradigms based on both traditional real-time computing and artificial intelligence. DAA consists of distributed agents that are classified into two categories: reactive and cognitive. Reactive agents can be implemented directly in Ada to meet hard real-time requirements and be deployed on on-board embedded processors. A traditional real-time computing methodology under consideration is the rate monotonic theory that can guarantee schedulability based on analytical methods. AI techniques under consideration for reactive agents are approximate or anytime reasoning that can be implemented using Bayesian belief networks as in Guardian. Cognitive agents are traditional expert systems that can be implemented in ART-Ada to meet soft real-time requirements. During the initial design of cognitive agents, it is critical to consider the migration path that would allow initial deployment on ground-based workstations with eventual deployment on on-board processors. ART-Ada technology enables this migration while Lisp-based technologies make it difficult if not impossible. In addition to reactive and cognitive agents, a meta-level agent would be needed to coordinate multiple agents and to provide meta-level control.
Aviation is one of the critical modes of our national transportation system. As such, it is essential that new technologies be continually developed to ensure that a safe mode of transportation becomes even safer in the future. The NASA Aviation Safety Program (AvSP) is managing the development of new technologies and interventions aimed at reducing the fatal aviation accident rate by a factor of 5 by year 2007 and by a factor of 10 by year 2022. A portfolio assessment is currently being conducted to determine the projected impact that the new technologies and/or interventions may have on reducing aviation safety system risk. This paper reports on advanced risk analytics that combine the use of a human error taxonomy, probabilistic Bayesian Belief Networks, and case-based scenarios to assess a relative risk intensity metric. A sample case is used for illustrative purposes.
In-flight loss of control (LOC) is currently the leading cause of fatal accidents based on various commercial aircraft accident statistics. As the Next Generation Air Transportation System (NextGen) emerges, new contributing factors leading to LOC are anticipated. The NASA Aviation Safety Program (AvSP), along with other aviation agencies and communities are actively developing safety products to mitigate the LOC risk. This paper discusses the approach used to construct a generic integrated LOC accident framework (LOCAF) model based on a detailed review of LOC accidents over the past two decades. The LOCAF model is comprised of causal factors from the domain of human factors, aircraft system component failures, and atmospheric environment. The multiple interdependent causal factors are expressed in an Object-Oriented Bayesian belief network. In addition to predicting the likelihood of LOC accident occurrence, the system-level integrated LOCAF model is able to evaluate the impact of new safety technology products developed in AvSP. This provides valuable information to decision makers in strategizing NASA's aviation safety technology portfolio. The focus of this paper is on the analysis of human causal factors in the model, including the contributions from flight crew and maintenance workers. The Human Factors Analysis and Classification System (HFACS) taxonomy was used to develop human related causal factors. The preliminary results from the baseline LOCAF model are also presented.
Aviation is a problem domain characterized by a high level of system complexity and uncertainty. Safety risk analysis in such a domain is especially challenging given the multitude of operations and diverse stakeholders. The Federal Aviation Administration (FAA) projects that by 2025 air traffic will increase by more than 50 percent with 1.1 billion passengers a year and more than 85,000 flights every 24 hours contributing to further delays and congestion in the sky (Circelli, 2011). This increased system complexity necessitates the application of structured safety risk analysis methods to understand and eliminate where possible, reduce, and/or mitigate risk factors. The use of expert judgments for probabilistic safety analysis in such a complex domain is necessary especially when evaluating the projected impact of future technologies, capabilities, and procedures for which current operational data may be scarce. Management of an R&D product portfolio in such a dynamic domain needs a systematic process to elicit these expert judgments, process modeling results, perform sensitivity analyses, and efficiently communicate the modeling results to decision makers. In this paper a case study focusing on the application of an R&D portfolio of aeronautical products intended to mitigate aircraft Loss of Control (LOC) accidents is presented. In particular, the knowledge elicitation process with three subject matter experts who contributed to the safety risk model is emphasized. The application and refinement of a verbal-numerical scale for conditional probability elicitation in a Bayesian Belief Network (BBN) is discussed. The preliminary findings from this initial step of a three-part elicitation are important to project management practitioners as they illustrate the vital contribution of systematic knowledge elicitation in complex domains.
The programs within NASA's Aeronautics Research Mission Directorate (ARMD) conduct research and development to improve the national air transportation system so that Americans can travel as safely as possible. NASA aviation safety systems analysis personnel support various levels of ARMD management in their fulfillment of system analysis and technology prioritization as defined in the agency's program and project requirements. This paper provides a framework for the assessment of aviation safety research and technology portfolios that includes metrics such as projected impact on current and future safety, technical development risk and implementation risk. The paper also contains methods for presenting portfolio analysis and aviation safety Bayesian Belief Network (BBN) output results to management using bubble charts and quantitative decision analysis techniques.