Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “bayesian optimization”

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 793 records · Page 44

PyOED: An Extensible Suite for Data Assimilation and Model-Constrained Optimal Design of Experiments

This article describes PyOED, a highly extensible scientific package that enables developing and testing model-constrained optimal experimental design (OED) for inverse problems. Specifically, PyOED aims to be a comprehensive Python toolkit for model-constrained OED. The package targets scientists and researchers interested in understanding the details of OED formulations and approaches. It is also meant to enable researchers to experiment with standard and innovative OED technologies with a wide range of test problems (e.g., simulation models). OED, inverse problems (e.g., Bayesian inversion), and data assimilation (DA) are closely related research fields, and their formulations overlap significantly. Thus, PyOED is continuously being expanded with a plethora of Bayesian inversion, DA, and OED methods as well as new scientific simulation models, observation error models, and observation operators. These pieces are added such that they can be permuted to enable testing OED methods in various settings of varying complexities. The PyOED core is completely written in Python and utilizes the inherent object-oriented capabilities; however, the current version of PyOED is meant to be extensible rather than scalable. Specifically, PyOED is developed to “enable rapid development and benchmarking of OED methods with minimal coding effort and to maximize code reutilization.” This article provides a brief description of the PyOED layout and philosophy and provides a set of exemplary test cases and tutorials to demonstrate the potential of the package.

97 MATHEMATICS AND COMPUTING↗

Role of the likelihood for elastic scattering uncertainty quantification

In the last decade, uncertainty quantification (UQ) for optical model potentials (OMPs) has become a focal point for nuclear reaction theory, and several competing approaches for OMP UQ have recently been developed. Here, we clarify recent efforts to compare frequentist and Bayesian approaches in the context of OMP UQ [G. B. King et al., Phys. Rev. Lett. 122, 232502 (2019)]. We replicate a portion of that OMP UQ study but use independent statistical tools. Specifically, we compare two methods for OMP parameter inference from elastic scattering data: the Levenberg-Marquardt algorithm for χ 2 minimization on one hand and Markov chain Monte Carlo (MCMC) sampling on the other. Separately, we assess the common practice of using a renormalized likelihood (χ 2 /N), N being the number of data points, instead of the canonical weighted-least-squares likelihood (χ 2 ), as a way of accounting for unknown data correlations. Here, we show that for a generic linear model and for a five-parameter OMP analysis, frequentist and uniform-prior Bayesian approaches recover the same optimum and uncertainty estimates—not systematically larger uncertainties for the Bayesian approach, as was concluded in G. B. King et al., Phys. Rev. Lett. 122, 232502 (2019). Further, we show that if an additional, near-degenerate parameter is introduced into the same OMP analysis such that the parameter posterior becomes non-Gaussian, then covariance-based estimates of uncertainty become unreliable. Finally, we show that regardless of optimization approach, if χ 2 /N is used for the likelihood, the resulting parametric uncertainties increase by $\sqrt{N}$, and that this is responsible for the conclusions drawn in the revisited study. Based on our replication results, we find that a fortuitous cancellation of unreported errors and the renormalization factor can lead to improvement in empirical coverages, as was the case in the original comparative study. We emphasize that developing and applying a realistic likelihood function is an essential task in a UQ analysis, and that several recent UQ studies that employed a renormalized likelihood (i.e., including a 1/N factor) may have yielded unrealistically large uncertainties for elastic-scattering observables. If the parameter posterior deviates from multivariate-normal, a sampling-based approach like MCMC has a clear advantage over methods that assume the Laplace approximation holds. We note that empirical coverage can serve as an important internal check for the analyst whose model or data may have additional, unaccounted-for uncertainties.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

A Bayesian Belief Network of Threat Anticipation and Terrorist Motivations

Recent events highlight the need for efficient tools for anticipating the threat posed by terrorists, whether individual or groups. Antiterrorism includes fostering awareness of potential threats, deterring aggressors, developing security measures, planning for future events, halting an event in process, and ultimately mitigating and managing the consequences of an event. To analyze such components, one must understand various aspects of threat elements like physical assets and their economic and social impacts. To this aim, we developed a three-layer Bayesian belief network (BBN) model that takes into consideration the relative threat of an attack against a particular asset (physical layer) as well as the individual psychology and motivations that would induce a person to either act alone or join a terrorist group and commit terrorist acts (social and economic layers). After researching the many possible motivations to become a terrorist, the main factors are compiled and sorted into categories such as initial and personal indicators, exclusion factors, and predictive behaviors. Assessing such threats requires combining information from disparate data sources most of which involve uncertainties. BBN combines these data in a coherent, analytically defensible, and understandable manner. The developed BBN model takes into consideration the likelihood and consequence of a threat in order to draw inferences about the risk of a terrorist attack so that mitigation efforts can be optimally deployed. The model is constructed using a network engineering process that treats the probability distributions of all the BBN nodes within the broader context of the system development process.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Holocene paleoclimate change in the western US: The importance of chronology in discerning patterns and drivers

We report sediment in lakes and meadows forms a powerful archive that can be used to reconstruct environmental change through time. Reconstructions of lake level, of chemical, biological, and hydrological conditions, and of surrounding vegetation provide detailed information about past climate conditions, both locally and regionally. Indeed, most of our current knowledge of centennial- to millennial-scale climate variability in the arid western United States, where information about past hydroclimate is particularly important, comes from such sediment-based reconstructions. The pressing need for robust, precise predictions of future conditions is a significant motivation for paleoclimate science, and current research questions frequently require Holocene reconstructions to be resolved at sub-centennial timescales. Increasingly, regional syntheses seek to identify synoptic-scale patterns similar to those defined from modern observations (seasonal, interannual, multi-decadal, etc.) or to compare with the output of climate model simulations. However, the age control on existing records, especially those more than about 20 years old, is often sufficient only for millennial-scale interpretation. Here we assess the age control for 84 published and unpublished records from lakes and meadows in the Great Basin, California, and desert southwest, and use Bayesian modeling to evaluate the 95% uncertainty ranges for the 42 best-dated records. In the Late Holocene, about half of the 42 records have <400-year mean uncertainty ranges; however, high-precision age control is especially critical for young records, used to develop an accurate understanding of a proxy’s response to known climate variations. In the Middle Holocene, records vary from 400 to >800-year mean uncertainty and records of the Early Holocene have 600- to >1400-year mean uncertainty ranges. We find that the largest control on modeled uncertainties is dating density, with at least 2 dates/kyr being optimal and suggest obtaining “range-finder” dates at the onset of a study to better predict the total number of dates needed for an adequate age model. Such a density avoids a commonly observed phenomenon of significant peaks in uncertainty arising in gaps between age control points. Analysis of the uncertainties associated with proxy shifts reveal that more than half are >400 years. Although such large uncertainties currently prevent sub-centennial interpretations in most cases, increased dating density, strategic use of limited funds (including budgeting for a 2 date/kyr minimum at the proposal stage), construction of age-depth models with Bayesian methods, and critical evaluation of chronological uncertainty will shed light on past climate variability at finer timescales, enhancing our understanding of global and regional drivers of western U.S. climate.

54 ENVIRONMENTAL SCIENCES↗

An Adaptive Sparse-Grid High-Order Stochastic Collocation Method for Bayesian Inference in Groundwater Reactive Transport Modeling

Although Bayesian analysis has become vital to the quantification of prediction uncertainty in groundwater modeling, its application has been hindered due to the computational cost associated with numerous model executions needed for exploring the posterior probability density function (PPDF) of model parameters. This is particularly the case when the PPDF is estimated using Markov Chain Monte Carlo (MCMC) sampling. In this study, we develop a new approach that improves computational efficiency of Bayesian inference by constructing a surrogate system based on an adaptive sparse-grid high-order stochastic collocation (aSG-hSC) method. Unlike previous works using first-order hierarchical basis, we utilize a compactly supported higher-order hierar- chical basis to construct the surrogate system, resulting in a significant reduction in the number of computational simulations required. In addition, we use hierarchical surplus as an error indi- cator to determine adaptive sparse grids. This allows local refinement in the uncertain domain and/or anisotropic detection with respect to the random model parameters, which further improves computational efficiency. Finally, we incorporate a global optimization technique and propose an iterative algorithm for building the surrogate system for the PPDF with multiple significant modes. Once the surrogate system is determined, the PPDF can be evaluated by sampling the surrogate system directly with very little computational cost. The developed method is evaluated first using a simple analytical density function with multiple modes and then using two synthetic groundwater reactive transport models. The groundwater models represent different levels of complexity; the first example involves coupled linear reactions and the second example simulates nonlinear ura- nium surface complexation. The results show that the aSG-hSC is an effective and efficient tool for Bayesian inference in groundwater modeling in comparison with conventional MCMC sim- ulations. The computational efficiency is expected to be more beneficial to more computational expensive groundwater problems.

Zhang, Guannan↗

Human limits in machine learning: prediction of potato yield and disease using soil microbiome data

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.

Aghdam, Rosa↗

SICK: THE SPECTROSCOPIC INFERENCE CRANK

There exists an inordinate amount of spectral data in both public and private astronomical archives that remain severely under-utilized. The lack of reliable open-source tools for analyzing large volumes of spectra contributes to this situation, which is poised to worsen as large surveys successively release orders of magnitude more spectra. In this article I introduce sick, the spectroscopic inference crank, a flexible and fast Bayesian tool for inferring astrophysical parameters from spectra. sick is agnostic to the wavelength coverage, resolving power, or general data format, allowing any user to easily construct a generative model for their data, regardless of its source. sick can be used to provide a nearest-neighbor estimate of model parameters, a numerically optimized point estimate, or full Markov Chain Monte Carlo sampling of the posterior probability distributions. This generality empowers any astronomer to capitalize on the plethora of published synthetic and observed spectra, and make precise inferences for a host of astrophysical (and nuisance) quantities. Model intensities can be reliably approximated from existing grids of synthetic or observed spectra using linear multi-dimensional interpolation, or a Cannon-based model. Additional phenomena that transform the data (e.g., redshift, rotational broadening, continuum, spectral resolution) are incorporated as free parameters and can be marginalized away. Outlier pixels (e.g., cosmic rays or poorly modeled regimes) can be treated with a Gaussian mixture model, and a noise model is included to account for systematically underestimated variance. Combining these phenomena into a scalar-justified, quantitative model permits precise inferences with credible uncertainties on noisy data. I describe the common model features, the implementation details, and the default behavior, which is balanced to be suitable for most astronomical applications. Using a forward model on low-resolution, high signal-to-noise ratio spectra of M67 stars reveals atomic diffusion processes on the order of 0.05 dex, previously only measurable with differential analysis techniques in high-resolution spectra. sick is easy to use, well-tested, and freely available online through GitHub under the MIT license.

79 ASTRONOMY AND ASTROPHYSICS↗

The Sixth Data Release of the Radial Velocity Experiment (Rave). II. Stellar Atmospheric Parameters, Chemical Abundances, and Distances

We present part 2 of the sixth and final Data Release (DR6) of the Radial Velocity Experiment (Rave), a magnitude-limited (9<I<12) spectroscopic survey of Galactic stars randomly selected in Earth’s southern hemisphere. The Rave medium-resolution spectra (R ∼ 7500) cover the Ca triplet region (8410–8795 Å) and span the complete time frame from the start of Rave observations on 2003 April 12 to their completion on 2013 April 4. In the second of two publications, we present the data products derived from 518,387 observations of 451,783 unique stars using a suite of advanced reduction pipelines focusing on stellar atmospheric parameters, in particular purely spectroscopically derived stellar atmospheric parameters (T{sub eff}, logg, and the overall metallicity), enhanced stellar atmospheric parameters inferred via a Bayesian pipeline using Gaia DR2 astrometric priors, and asteroseismically calibrated stellar atmospheric parameters for giant stars based on asteroseismic observations for 699 K2 stars. In addition, we provide abundances of the elements Fe, Al, and Ni, as well as an overall [α/Fe] ratio obtained using a new pipeline based on the GAUGUIN optimization method that is able to deal with variable signal-to-noise ratios. The Rave DR6 catalogs are cross-matched with relevant astrometric and photometric catalogs, and are complemented by orbital parameters and effective temperatures based on the infrared flux method. The data can be accessed via the Rave website (http://rave-survey.org) or the Vizier database.

79 ASTRONOMY AND ASTROPHYSICS↗

Uncertainty quantification and reliability assessment for intermodal freight transportation

Intermodal freight optimization models support cost-effective, low-emission, and timely goods movement by coordinating trucks, rail, and barges. These models determine optimal flows, routing, and modal switches while respecting infrastructure and operational constraints. However, their real-world utility is often undermined by pervasive uncertainties-such as fluctuating transportation costs and emissions, variable terminal capacities, and uncertain freight demand-that distort key performance outcomes, including total system cost, carbon footprint, and transit time reliability. This study presents a structured framework for quantifying uncertainty in intermodal freight transportation (IFT) optimization. The framework evaluates how input uncertainty affects system performance and reliability, a critical need for ensuring that model-based decisions remain robust under real-world variability, especially amid volatile fuel prices, shifting demand, and growing disruptions. It integrates three complementary methods: (1) Sobol-based global sensitivity analysis to identify influential parameters affecting cost, emissions, and transit time, (2) Monte Carlo-based capacity perturbation analysis to assess robustness under probabilistic facility disruptions, and (3) Monte Carlo filtering with Bayesian inference to detect threshold-based performance vulnerabilities. The results highlight diesel truck unit cost as the dominant driver of variability. To improve system resilience, planners should prioritize uncertainty in fuel-related parameters when designing intermodal strategies.

Intermodal freight transportation↗

NONPARAMETRIC BAYESIAN ESTIMATION OF PERIODIC LIGHT CURVES

Many astronomical phenomena exhibit patterns that have periodic behavior. An important step when analyzing data from such processes is the problem of identifying the period: estimating the period of a periodic function based on noisy observations made at irregularly spaced time points. This problem is still a difficult challenge despite extensive study in different disciplines. This paper makes several contributions toward solving this problem. First, we present a nonparametric Bayesian model for period finding, based on Gaussian Processes (GPs), that does not make assumptions on the shape of the periodic function. As our experiments demonstrate, the new model leads to significantly better results in period estimation especially when the light curve does not exhibit sinusoidal shape. Second, we develop a new algorithm for parameter optimization for GP which is useful when the likelihood function is very sensitive to the parameters with numerous local minima, as in the case of period estimation. The algorithm combines gradient optimization with grid search and incorporates several mechanisms to overcome the high computational complexity of GP. Third, we develop a novel approach for using domain knowledge, in the form of a probabilistic generative model, and incorporate it into the period estimation algorithm. Experimental results validate our approach showing significant improvement over existing methods.

79 ASTRONOMY AND ASTROPHYSICS↗

PRIMORDIAL GRAVITATIONAL WAVE DETECTABILITY WITH DEEP SMALL-SKY COSMIC MICROWAVE BACKGROUND EXPERIMENTS

We use the Bayesian estimation on direct T - Q - U cosmic microwave background (CMB) polarization maps to forecast errors on the tensor-to-scalar power ratio r, and hence on primordial gravitational waves, as a function of sky coverage f{sub sky}. This map-based likelihood filters the information in the pixel-pixel space into the optimal combinations needed for r detection for cut skies, providing enhanced information over a first-step linear separation into a combination of E, B, and mixed modes, and ignoring the latter. With current computational power and for typical resolutions appropriate for r detection, the large matrix inversions required are accurate and fast. Our simulations explore two classes of experiments, with differing bolometric detector numbers, sensitivities, and observational strategies. One is motivated by a long duration balloon experiment like Spider, with pixel noise {proportional_to}{radical}(f{sub sky}) for a specified observing period. This analysis also applies to ground-based array experiments. We find that, in the absence of systematic effects and foregrounds, an experiment with Spider-like noise concentrating on f{sub sky} {approx} 0.02-0.2 could place a 2{sigma}{sub r} Almost-Equal-To 0.014 boundary ({approx}95% confidence level), which rises to 0.02 with an l-dependent foreground residual left over from an assumed efficient component separation. We contrast this with a Planck-like fixed instrumental noise as f{sub sky} varies, which gives a Galaxy-masked (f{sub sky} = 0.75) 2{sigma}{sub r} Almost-Equal-To 0.015, rising to Almost-Equal-To 0.05 with the foreground residuals. Using as the figure of merit the (marginalized) one-dimensional Shannon entropy of r, taken relative to the first 2003 WMAP CMB-only constraint, gives -2.7 bits from the 2012 WMAP9+ACT+SPT+LSS data, and forecasts of -6 bits from Spider (+ Planck); this compares with up to -11 bits for CMBPol, COrE, and PIXIE post-Planck satellites and -13 bits for a perfectly noiseless cosmic variance limited experiment. We thus confirm the wisdom of the current strategy for r detection of deeply probed patches covering the f{sub sky} minimum-error trough with balloon and ground experiments.

79 ASTRONOMY AND ASTROPHYSICS↗

Estimating building simulation parameters via Bayesian structure learning

Many key building design policies are made using sophisticated computer simulations such as EnergyPlus (E+), the DOE flagship whole-building energy simulation engine. E+ and other sophisticated computer simulations have several major problems. The two main issues are 1) gaps between the simulation model and the actual structure, and 2) limitations of the modeling engine's capabilities. Currently, these problems are addressed by having an engineer manually calibrate simulation parameters to real world data or using algorithmic optimization methods to adjust the building parameters. However, some simulations engines, like E+, are computationally expensive, which makes repeatedly evaluating the simulation engine costly. This work explores addressing this issue by automatically discovering the simulation's internal input and output dependencies from ~20 Gigabytes of E+ simulation data, future extensions will use ~200 Terabytes of E+ simulation data. The model is validated by inferring building parameters for E+ simulations with ground truth building parameters. Our results indicate that the model accurately represents parameter means with some deviation from the means, but does not support inferring parameter values that exist on the distribution's tail.

Edwards, Richard E.↗

Estimating Building Simulation Parameters via Bayesian Structure Learning

Many key building design policies are made using sophisticated computer simulations such as EnergyPlus (E+), the DOE flagship whole-building energy simulation engine. E+ and other sophisticated computer simulations have several major problems. The two main issues are 1) gaps between the simulation model and the actual structure, and 2) limitations of the modeling engine's capabilities. Currently, these problems are addressed by having an engineer manually calibrate simulation parameters to real world data or using algorithmic optimization methods to adjust the building parameters. However, some simulations engines, like E+, are computationally expensive, which makes repeatedly evaluating the simulation engine costly. This work explores addressing this issue by automatically discovering the simulation's internal input and output dependencies from 20 Gigabytes of E+ simulation data, future extensions will use 200 Terabytes of E+ simulation data. The model is validated by inferring building parameters for E+ simulations with ground truth building parameters. Our results indicate that the model accurately represents parameter means with some deviation from the means, but does not support inferring parameter values that exist on the distribution's tail.

Energy Modeling↗

Multi-fidelity Gaussian process regression for prediction of random fields

We propose a new multi-fidelity Gaussian process regression (GPR) approach for prediction of random fields based on observations of surrogate models or hierarchies of surrogate models. Our method builds upon recent work on recursive Bayesian techniques, in particular recursive co-kriging, and extends it to vector-valued fields and various types of covariances, including separable and non-separable ones. The framework we propose is general and can be used to perform uncertainty propagation and quantification in model-based simulations, multi-fidelity data fusion, and surrogate-based optimization. We demonstrate the effectiveness of the proposed recursive GPR techniques through various examples. Specifically, we study the stochastic Burgers equation and the stochastic Oberbeck–Boussinesq equations describing natural convection within a square enclosure. In both cases we find that the standard deviation of the Gaussian predictors as well as the absolute errors relative to benchmark stochastic solutions are very small, suggesting that the proposed multi-fidelity GPR approaches can yield highly accurate results.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Shallow snowpack inhibits soil respiration in sagebrush steppe through multiple biotic and abiotic mechanisms

Abstract In sagebrush steppe, snowpack may govern soil respiration through its effect on multiple abiotic and biotic factors. Across the Intermountain West of the United States, snowpack has been declining for decades and is projected to decline further over the next century, making the response of soil respiration to snowpack a potentially important factor in the ecosystem carbon cycle. In this study, we evaluated the direct and indirect roles of the snowpack in driving soil respiration in sagebrush steppe ecosystems by taking advantage of highway snowfences in Wyoming to manipulate snowpack. An important contribution of this study is the use of Bayesian modeling to quantify the effects of soil moisture and temperature on soil respiration across a wide range of conditions from frozen to hot and dry, while simultaneously accounting for biotic factors (e.g., vegetation cover, root density, and microbial biomass and substrate‐use diversity) affected by snowpack. Elevated snow depth increased soil temperature (in the winter) and moisture (winter and spring), and was associated with reduced vegetation cover and microbial biomass carbon. Soil respiration showed an exponential increase with temperature, with a temperature sensitivity that decreased with increasing seasonal temperature ( Q 10 = 4.3 [winter], 2.3 [spring], and 1.7 [summer]); frozen soils were associated with unrealistic Q 10 ≈ 7989 due to the liquid‐to‐ice transition of soil water. Soil respiration was sensitive to soil water content; predicted respiration under very dry conditions was less than 10% of respiration under moist conditions. While higher vegetation cover increased soil respiration, this was not due to increased root density, and may reflect differences in litter inputs. Microbial substrate‐use diversity was negatively related to reference respiration (i.e., respiration rate at a reference temperature and optimal soil moisture), although the mechanism remains unclear. This study indicates that soil respiration is inhibited by shallow snowpack through multiple mechanisms; thus, future decreases in snowpack across the sagebrush steppe have the potential to reduce losses of soil C, potentially affecting regional carbon balance.

59 BASIC BIOLOGICAL SCIENCES↗

Probing the primordial power spectra with inflationary priors

We investigate constraints on power spectra of the primordial curvature and tensor perturbations with priors based on single-field slow-roll inflation models. The Hubble slow-roll parameters are included in cosmological parameters and the primordial power spectra are generated using the inflationary flow equations. Using data from recent observations of CMB and several measurements of geometrical distances in the late Universe, we perform Bayesian parameter estimation and model selection for models that have separate priors on the slow-roll parameters. The same analysis is also performed adopting the standard parameterization of the primordial power spectra. We confirmed that the scale-invariant Harrison-Zel'dovich spectrum is disfavored with more significance than previous studies. While current observations appear to be optimally modeled with some simple models of single-field slow-roll inflation, data is not enough constraining to distinguish these models.

79 ASTRONOMY AND ASTROPHYSICS↗

Cyclotron radiation emission spectroscopy of electrons from tritium $β$ decay and 83m Kr internal conversion

Project 8 has developed a novel technique, cyclotron radiation emission spectroscopy (CRES), for direct neutrino mass measurements. A CRES-based experiment on the beta spectrum of tritium has been carried out in a small-volume apparatus. Here, we provide a detailed account of the experiment, focusing on systematic effects and analysis techniques. In a Bayesian (frequentist) analysis, we measure the tritium endpoint as ${18}$ ${553}_{—19}^{+18}$ (${18}$ ${548}_{—19}^{+19}$) eV and set upper limits of 155 (152) eV (90% C.L.) on the neutrino mass. No background events are observed beyond the endpoint in 82 days of running. We also demonstrate an energy resolution of 1.66 ± 0.19 eV in a resolution-optimized magnetic trap configuration by measuring 83m Kr 17.8-keV internal-conversion electrons. These measurements establish CRES as a low-background, high-resolution technique with the potential to advance neutrino mass sensitivity

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A Bayesian Learning Approach to Wireless Outdoor Heatmap Construction using Deep Gaussian Process

We present a novel Bayesian learning approach to outdoor radio heatmap construction utilizing deep Gaussian process (GP). The proposed approach employs a two-layer hierarchy which consists of two cascaded Gaussian processes that are capable of modeling more complex input-output relations than standard single-layer Gaussian processes. Since deriving the exact model likelihood is challenging, a lower bound is optimized instead so that gradient descent-based methods can be performed to find out the optimal model parameters. Typically, inducing points are used in GPs to facilitate low-rank approximation of covariance (kernel) matrices for computation speedup. However, the inaccuracy induced by inducing points can accumulate when stacking multiple layers of GP which may hinder the performance of deep GP. Moreover, since inducing points need to be learned, having them at all layers of deep GP also incurs computational burden. To overcome the above challenges, in contrast to the canonical deep GP model, we use a modified architecture where a full standard GP resides in the first layer and inducing points are only introduced for the second layer. This modified architecture strikes a balance between model accuracy and training complexity. In the proposed model, the noise parameter of the first GP layer is also eliminated to improve the training efficiency as the noise parameter at the output of the second layer suffices to model the uncertainty in the output. The proposed approach is evaluated on real-world datasets, in the form of location-Received Signal Strength (RSS) pairs, collected from the Platform for Open Wireless Data-driven Experimental Research (POWDER) located at the campus of the University of Utah. Experiment results show that the proposed approach can achieve smaller prediction errors on various training and testing data configurations than DNN-based and GP-based methods.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗