Hybrid SAFE-GMM approach for predictive modeling of guided wave propagation in layered media
Not provided.
Engineering↗
SEARCH · Engineering Papers
Results for “GMM”
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.
PSD GMM Trainer
Code takes a data file of input features of a set of digitized pulses and converts the file to three vectors describing a two-component Gaussian mixture model of the input pulses. These three vectors can then be applied mathematically to a set of input feature data of digitized pulses to generate soft-scores describing the likelihood of each pulse's membership in each of the two Gaussian modeled components. The operation of the code also displays graphics to the user for interpreting the output vectors and distribution of the soft-scores of the input data.
Blair, BrentonS↗
Th1-skewed tissue responses to a mycolyl glycolipid in mycobacteria-infected rhesus macaques
Highlights: •Glucose monomycolate (GMM) is a marker glycolipid for active tuberculosis. •Tissue responses to GMM involved up-regulation of Th1-attracting chemokines. •Th1-skewed local responses were mounted at the GMM-injected tissue. -- Abstract: Trehalose 6,6′-dimycolate (TDM) is a major glycolipid of the cell wall of mycobacteria with remarkable adjuvant functions. To avoid detection by the host innate immune system, invading mycobacteria down-regulate the expression of TDM by utilizing host-derived glucose as a competitive substrate for their mycolyltransferases; however, this enzymatic reaction results in the concomitant biosynthesis of glucose monomycolate (GMM) which is recognized by the acquired immune system. GMM-specific, CD1-restricted T cell responses have been detected in the peripheral blood of infected human subjects and monkeys as well as in secondary lymphoid organs of small animals, such as guinea pigs and human CD1-transgenic mice. Nevertheless, it remains to be determined how tissues respond at the site where GMM is produced. Here we found that rhesus macaques vaccinated with Mycobacterium bovis bacillus Calmette–Guerin mounted a chemokine response in GMM-challenged skin that was favorable for recruiting T helper (Th)1 T cells. Indeed, the expression of interferon-γ, but not Th2 or Th17 cytokines, was prominent in the GMM-injected tissue. The GMM-elicited tissue response was also associated with the expression of monocyte/macrophage-attracting CC chemokines, such as CCL2, CCL4 and CCL8. Furthermore, the skin response to GMM involved the up-regulated expression of granulysin and perforin. Given that GMM is produced primarily by pathogenic mycobacteria proliferating within the host, the Th1-skewed tissue response to GMM may function efficiently at the site of infection.
60 APPLIED LIFE SCIENCES↗
Efficient Speaker Verification Using Gaussian Mixture Model Component Clustering
In speaker verification (SV) systems that employ a support vector machine (SVM) classifier to make decisions on a supervector derived from Gaussian mixture model (GMM) component mean vectors, a significant portion of the computational load is involved in the calculation of the a posteriori probability of the feature vectors of the speaker under test with respect to the individual component densities of the universal background model (UBM). Further, the calculation of the sufficient statistics for the weight, mean, and covariance parameters derived from these same feature vectors also contribute a substantial amount of processing load to the SV system. In this paper, we propose a method that utilizes clusters of GMM-UBM mixture component densities in order to reduce the computational load required. In the adaptation step we score the feature vectors against the clusters and calculate the a posteriori probabilities and update the statistics exclusively for mixture components belonging to appropriate clusters. Each cluster is a grouping of multivariate normal distributions and is modeled by a single multivariate distribution. As such, the set of multivariate normal distributions representing the different clusters also form a GMM. This GMM is referred to as a hash GMM which can be considered to a lower resolution representation of the GMM-UBM. The mapping that associates the components of the hash GMM with components of the original GMM-UBM is referred to as a shortlist. This research investigates various methods of clustering the components of the GMM-UBM and forming hash GMMs. Of five different methods that are presented one method, Gaussian mixture reduction as proposed by Runnall's, easily outperformed the other methods. This method of Gaussian reduction iteratively reduces the size of a GMM by successively merging pairs of component densities. Pairs are selected for merger by using a Kullback-Leibler based metric. Using Runnal's method of reduction, we were able to achieve a factor of 2.77 reduction in a posteriori probability calculations with no loss in accuracy when the original UBM consisted of 256 component densities. When clustering was implemented with a 1024 component UBM, we achieved a computation reduction of 5 with no loss in accuracy and a reduction by a factor of 10 with less than 2.4% relative loss in accuracy.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Background based Gaussian mixture model lesion segmentation in PET
Purpose: Quantitative {sup 18}F-fluorodeoxyglucose positron emission tomography is limited by the uncertainty in lesion delineation due to poor SNR, low resolution, and partial volume effects, subsequently impacting oncological assessment, treatment planning, and follow-up. The present work develops and validates a segmentation algorithm based on statistical clustering. The introduction of constraints based on background features and contiguity priors is expected to improve robustness vs clinical image characteristics such as lesion dimension, noise, and contrast level. Methods: An eight-class Gaussian mixture model (GMM) clustering algorithm was modified by constraining the mean and variance parameters of four background classes according to the previous analysis of a lesion-free background volume of interest (background modeling). Hence, expectation maximization operated only on the four classes dedicated to lesion detection. To favor the segmentation of connected objects, a further variant was introduced by inserting priors relevant to the classification of neighbors. The algorithm was applied to simulated datasets and acquired phantom data. Feasibility and robustness toward initialization were assessed on a clinical dataset manually contoured by two expert clinicians. Comparisons were performed with respect to a standard eight-class GMM algorithm and to four different state-of-the-art methods in terms of volume error (VE), Dice index, classification error (CE), and Hausdorff distance (HD). Results: The proposed GMM segmentation with background modeling outperformed standard GMM and all the other tested methods. Medians of accuracy indexes were VE <3%, Dice >0.88, CE <0.25, and HD <1.2 in simulations; VE <23%, Dice >0.74, CE <0.43, and HD <1.77 in phantom data. Robustness toward image statistic changes (±15%) was shown by the low index changes: <26% for VE, <17% for Dice, and <15% for CE. Finally, robustness toward the user-dependent volume initialization was demonstrated. The inclusion of the spatial prior improved segmentation accuracy only for lesions surrounded by heterogeneous background: in the relevant simulation subset, the median VE significantly decreased from 13% to 7%. Results on clinical data were found in accordance with simulations, with absolute VE <7%, Dice >0.85, CE <0.30, and HD <0.81. Conclusions: The sole introduction of constraints based on background modeling outperformed standard GMM and the other tested algorithms. Insertion of a spatial prior improved the accuracy for realistic cases of objects in heterogeneous backgrounds. Moreover, robustness against initialization supports the applicability in a clinical setting. In conclusion, application-driven constraints can generally improve the capabilities of GMM and statistical clustering algorithms.
60 APPLIED LIFE SCIENCES↗
An optimized method to identify RR Lyrae stars in the SDSS×Pan-STARRS1 overlapping area using a bayesian generative technique
We present a method for selecting RR Lyrae (RRL) stars (or other types of variable stars) in the absence of a large number of multi-epoch data and light curve analyses. Our method uses color and variability selection cuts that are defined by applying a Gaussian Mixture Bayesian Generative Method (GMM) on 636 pre-identified RRL stars instead of applying the commonly used rectangular cuts. Specifically, our method selects 8115 RRL candidates (heliocentric distances < 70 kpc) using GMM color cuts from the Sloan Digital Sky Survey (SDSS) and GMM variability cuts from the Panoramic Survey Telescope and Rapid Response System 1 3π survey (PS1). Comparing our method with the Stripe 82 catalog of RRL stars shows that the efficiency and completeness levels of our method are ∼77% and ∼52%, respectively. Most contaminants are either non-variable main-sequence stars or stars in eclipsing systems. The method described here efficiently recovers known stellar halo substructures. It is expected that the current completeness and efficiency levels will further improve with the additional PS1 epochs (∼3 epochs per filter) that will be observed before the conclusion of the survey. A comparison between our efficiency and completeness levels using the GMM method to the efficiency and completeness levels using rectangular cuts that are commonly used yielded a significant increase in the efficiency level from ∼13% to ∼77% and an insignificant change in the completeness levels. Hence, we favor using the GMM technique in future studies. Although we develop it over the SDSS×PS1 footprint, the technique presented here would work well on any multi-band, multi-epoch survey for which the number of epochs is limited.
79 ASTRONOMY AND ASTROPHYSICS↗
A robust estimator of mutual information for deep learning interpretability
Abstract We develop the use of mutual information (MI), a well-established metric in information theory, to interpret the inner workings of deep learning (DL) models. To accurately estimate MI from a finite number of samples, we present GMM-MI (pronounced ‘Jimmie’), an algorithm based on Gaussian mixture models that can be applied to both discrete and continuous settings. GMM-MI is computationally efficient, robust to the choice of hyperparameters and provides the uncertainty on the MI estimate due to the finite sample size. We extensively validate GMM-MI on toy data for which the ground truth MI is known, comparing its performance against established MI estimators. We then demonstrate the use of our MI estimator in the context of representation learning, working with synthetic data and physical datasets describing highly non-linear processes. We train DL models to encode high-dimensional data within a meaningful compressed (latent) representation, and use GMM-MI to quantify both the level of disentanglement between the latent variables, and their association with relevant physical quantities, thus unlocking the interpretability of the latent representation. We make GMM-MI publicly available in this GitHub repository.
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
Green micromachining of ceramics using tungsten carbide micro-endmills
This paper presents an experimental analysis on micromachining of green-state silicon carbide (SiC) and aluminum nitride (AlN) using uncoated tungsten carbide (WC) micro-endmills. Although the unique properties of ceramics make them ideal materials for many applications involving micro-scale features, the industrial adaptation of ceramics at the micro-scale has been hindered due to the lack of viable ceramic micro-manufacturing techniques. Green micromachining (GMM), where micro-scale features are created by micromachining green-state ceramics that contain ceramic particles and a polymer binder, offers a potential solution for this problem by affording tremendous improvements in micromachinability with respect to sintered ceramics. After the completion of GMM, the green parts are debound and sintered to obtain the final component with the desired microscale features. In this work, we evaluated GMM characteristics of powder-injection molded green-state SiC and AlN to correlate process conditions and green-material compositions with the micromachining forces, quality, and tool wear. An experimental study that involves full immersion micromilling tests with 254 μm diameter micro-endmills at different cutting speeds and feed rates is conducted. For each of the two materials, two binder states (with and without wax) and two powder states (with micro particles only, or with both micro- and nanoparticles) are investigated. Green micromachining forces, specific energies and the resulting surface roughness are analyzed quantitatively. A qualitative analysis of burr formation and channel quality, as well as a preliminary study on micro-tool wear are also performed. Both the machining conditions and the material compositions are seen to have a strong effect on micro-machinability of green-state ceramics. Overall, we conclude that GMM can be a viable ceramic micro-manufacturing strategy if favorable (or optimal) micromachining and material parameters are identified through the presented (or a similar) micromachinability study.
36 MATERIALS SCIENCE↗
Host-region parameters for an adjustable model for crustal earthquakes to facilitate the implementation of the backbone approach to building ground-motion logic trees in probabilistic seismic hazard analysis
The backbone approach to constructing a ground-motion logic tree for probabilistic seismic hazard analysis (PSHA) can address shortcomings in the traditional approach of populating the branches with multiple existing, or potentially modified, ground-motion models (GMMs) by rendering more transparent the relationship between branch weights and the resulting distribution of predicted accelerations. To capture epistemic uncertainty in a tractable manner, there are benefits in building the logic tree through the application of successive adjustments for differences in source, path, and site characteristics between the host region of the selected backbone GMM and the target region for which the PSHA is being conducted. The implementation of this approach is facilitated by selecting a backbone GMM that is amenable to such host-to-target adjustments for individual source, path, and site characteristics. The NGA-West2 GMM of Chiou and Youngs (CY14) has been identified as a highly adaptable model for crustal seismicity that is well suited to such adjustments. Rather than using generic source, path, and site characteristics assumed appropriate for the host region, the final suite of adjusted GMMs for the target region will be better constrained if the host-region parameters are defined specifically on the basis of their compatibility with the CY14 backbone GMM. To this end, making use of a recently developed crustal shear-wave velocity profile consistent with CY14, we present an inversion of the model to estimate the key source and path parameters, namely the stress parameter and the anelastic attenuation. With these outputs, the effort in constructing a ground-motion logic tree for any PSHA dealing with crustal seismicity can be focused primarily on the estimation of the target-region characteristics and their associated uncertainties. The inversion procedure can also be adapted for any application in which different constraints might be relevant.
Stafford, Peter J.↗
Tailoring Microbial Fitness Through Computational Steering and CRISPRi-Driven Robustness Regulation
The widespread application of genetically modified microorganisms (GMMs) across diverse sectors underscores the pressing need for robust strategies to mitigate the risks associated with their potential uncontrolled escape. This study merges computational modeling with CRISPR interference (CRISPRi) to refine GMM metabolic robustness. Utilizing ensemble modeling, we achieved high-throughput in silico screening for enzymatic targets susceptible to expression alterations. Translating these insights, we developed functional CRISPRi, boosting fitness control via multiplexed gene knockdown. Our method, enhanced by an insulator-improved gRNA structure and an off-switch circuit controlling a compact Cas12m, resulted in rationally engineered strains with escape frequencies below National Institutes of Health standards. The effectiveness of this approach was confirmed under various conditions, showcasing its ability for secure GMM management. This research underscores the resilience of microbial metabolism, strategically modifying key nodes to halt growth without provoking significant resistance, thereby enabling more reliable and precise GMM control. A record of this paper's transparent peer review process is included in the supplemental information.
59 BASIC BIOLOGICAL SCIENCES↗
Pulse discrimination with a Gaussian mixture model on an FPGA
A Gaussian Mixture Model (GMM) based machine learning algorithm has been applied to the problem of gamma/neutron pulse shape discrimination (PSD). The algorithm has been successfully implemented on a standard PC as well as a field programmable gate array (FPGA). We describe the GMM classifier and its implementation on these two different types of hardware. We compare the performance of the algorithm on these two platforms against each other, along with other standard techniques applied in PSD. Our results show that the FPGA-based GMM classifier outperforms the standard PSD techniques in terms of classification accuracy at low particle energy and executes more quickly than its CPU-based counterpart.
73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
Reducing computation in an i-vector speaker recognition system using a tree-structured universal background model
The majority of state-of-the-art speaker recognition systems (SR) utilize speaker models that are derived from an adapted universal background model (UBM) in the form of a Gaussian mixture model (GMM). This is true for GMM supervector systems, joint factor analysis systems, and most recently i-vector systems. In all of the identified systems, the posterior probabilities and sufficient statistics calculations represent a computational bottleneck in both enrollment and testing. We propose a multi-layered hash system, employing a tree-structured GMM–UBM which uses Runnalls’ Gaussian mixture reduction technique, in order to reduce the number of these calculations. Moreover, with this tree-structured hash, we can trade-off reduction in computation with a corresponding degradation of equal error rate (EER). As an example, we also reduce this computation by a factor of 15× while incurring less than 10% relative degradation of EER (or 0.3% absolute EER) when evaluated with NIST 2010 speaker recognition evaluation (SRE) telephone data.
42 ENGINEERING↗
Natural generalized mirage mediation
We present that in the supersymmetric scenario known as mirage mediation (MM), the soft SUSY breaking terms receive comparable anomaly-mediation and moduli-mediation contributions leading to the phenomenon of mirage unification. The simplest MM SUSY breaking models which are consistent with the measured Higgs mass and sparticle mass constraints are strongly disfavoured by fine-tuning considerations. However, while MM makes robust predictions for gaugino masses, the scalar sector is quite sensitive to specific mechanisms for moduli stabilization and potential uplifting. We suggest here a broader setup of generalized mirage mediation (GMM), where heretofore discrete parameters are allowed as continuous to better parametrize these other schemes. We find that natural SUSY spectra consistent with both the measured value of m h as well as LHC lower bounds on superpartner masses are then possible. We explicitly show that models generated from natural GMM may be beyond the reach of even high-luminosity LHC searches. In such a case, the proposed International Linear e + e - Collider (ILC) will be required for natural SUSY discovery via higgsino pair production reactions. We also outline prospects for detection of higgsino-like WIMPs from natural GMM.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
SU-G-JeP3-04: Estimating 4D CBCT from Prior Information and Extremely Limited Angle Projections Using Structural PCA and Weighted Free-Form Deformation
Purpose: To investigate the feasibility of using structure-based principal component analysis (PCA) motion-modeling and weighted free-form deformation to estimate on-board 4D-CBCT using prior information and extremely limited angle projections for potential 4D target verification of lung radiotherapy. Methods: A technique for lung 4D-CBCT reconstruction has been previously developed using a deformation field map (DFM)-based strategy. In the previous method, each phase of the 4D-CBCT was generated by deforming a prior CT volume. The DFM was solved by a motion-model extracted by global PCA and a free-form deformation (GMM-FD) technique, using data fidelity constraint and the deformation energy minimization. In this study, a new structural-PCA method was developed to build a structural motion-model (SMM) by accounting for potential relative motion pattern changes between different anatomical structures from simulation to treatment. The motion model extracted from planning 4DCT was divided into two structures: tumor and body excluding tumor, and the parameters of both structures were optimized together. Weighted free-form deformation (WFD) was employed afterwards to introduce flexibility in adjusting the weightings of different structures in the data fidelity constraint based on clinical interests. XCAT (computerized patient model) simulation with a 30 mm diameter lesion was simulated with various anatomical and respirational changes from planning 4D-CT to onboard volume. The estimation accuracy was evaluated by the Volume-Percent-Difference (VPD)/Center-of-Mass-Shift (COMS) between lesions in the estimated and “ground-truth” on board 4D-CBCT. Results: Among 6 different XCAT scenarios corresponding to respirational and anatomical changes from planning CT to on-board using single 30° on-board projections, the VPD/COMS for SMM-WFD was reduced to 10.64±3.04%/1.20±0.45mm from 21.72±9.24%/1.80±0.53mm for GMM-FD. Using 15° orthogonal projections, the VPD/COMS was further reduced to 1.91±0.86%/0.31±0.42mm based on SMM-WFD. Conclusion: Compared to GMM-FD technique, the SMM-WFD technique can substantially improve the 4D-CBCT estimation accuracy using extremely small scan angles to provide ultra-fast 4D verification. This work was supported by the National Institutes of Health under Grant No. R01-CA184173 and a research grant from Varian Medical Systems.
60 APPLIED LIFE SCIENCES↗
FORTÉ Machine Learning
Of the non-corrupted data collected by the Orbiting Experiment (forté) satellite’s Photo-Diode Detector during the year 2001, I estimate that 7.9% of 914 894 signals are noise. My result differs dramatically from Guillen’s estimate of 96%. To arrive at this estimate, I used Gaussian mixture model (GMM) clustering–unsupervised machine learning–to aggregate the wave forms into groups based on the absolute value of the lowest 25 positive frequency discrete Fourier transform coefficients. Then, I marked several of the groups as noise by inspecting a random sampling of wave forms from each group. Marking groups as either noise or non-noise is a supervised binary classification operation. After removing the signals in noise groups from further consideration, I clustered the remaining signals into families. Again, I used a GMM, but for the familial clustering I used a Non-Negative Matrix Factorization feature vector transform. The result was 9 distinct families of lightning signals, as well as a second stage of noise filtering. To efficiently represent the entirety of the signal space, I broke each family into deciles based on their distance from the family mean. In this case, distance means the log-likelihood based on the GMM. Signals in lower deciles are more similar in shape and amplitude to their family average. I took the top 200 samples from each decile of each group, resulting in 18 000 signals. These signal approximately represent the entirety of the forté observations. To represent outliers, I also kept a zoo of the 1000 signals furthest from any family’s average. All told, the resulting data set represents the forté data with a reduction of about 51:1. To allow synthesis of an arbitrarily large number of test signals, I also captured each family’s average signal and the time-sample covariance matrix over the signals in each family. Using these two pieces of information, I can synthesize new waveforms by using a Gaussian random realization from the family average and covariance matrix. I wrote a program to test the synthesis quality. The program shows me two signals on the screen, one synthesized and one randomly drawn from the data. I attempted to identify the synthesized signal. Although the synthesis is imperfect, in an A/B comparison I only correctly chose the synthesized signal 36% of the time.
97 MATHEMATICS AND COMPUTING↗
GPR Imaging for Deeply Buried Objects: A Comparative Study Based on Compositing of Scanning Frequencies and a Chirp Excitation Function
Compositing of ground penetrating radar (GPR) scans of differing frequencies have been found to produce cleaner images at depth using the Gaussian mixture model (GMM) feature of the expectation-maximization (EM) algorithm. GPR scans at various heights (“Stand Off”), as well as ground-based scans, have been studied. In this paper, we compare the GPR response from a chirp excitation function-based radar with the response from the EM GMM algorithm compositing process, using the same mix of frequencies. A chirp excitation pulse was found to be effective in delineating the defined buried object, but the resulting image is less sharp than the GMM EM method.
58 GEOSCIENCES↗
Contribution of Gaia Sausage to the Galactic Stellar Halo Revealed by K Giants and Blue Horizontal Branch Stars from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope, Sloan Digital Sky Survey, and Gaia
We explore the contribution of the Gaia Sausage to the stellar halo of the Milky Way by making use of a Gaussian mixture model (GMM) and applying it to halo star samples of Large Sky Area Multi-Object Fiber Spectroscopic Telescope K giants, Sloan Extension for Galactic Understanding and Exploration K giants, and Sloan Digital Sky Survey blue horizontal branch stars. The GMM divides the stellar halo into two parts, of which one represents a more metal-rich and highly radially biased component associated with an ancient, head-on collision referred to as the Gaia Sausage, and the other one is a more metal-poor and isotropic halo. A symmetric bimodal Gaussian is used to describe the distribution of spherical velocity of the Gaia Sausage, and we find that the mean absolute radial velocity of the two lobes decreases with the Galactocentric radius. We find that the Gaia Sausage contributes about 41%–74% of the inner (Galactocentric radius r gc < 30 kpc) stellar halo. The fraction of stars of the Gaia Sausage starts to decline beyond r gc ~ 25–30 kpc, and the outer halo is found to be significantly less influenced by the Gaia Sausage than the inner halo. After the removal of halo substructures found by integrals of motion, the contribution of the Gaia Sausage falls slightly within r gc ~ 25 kpc but is still as high as 30%–63%. Finally, we select several possible Sausage-related substructures consisting of stars on highly eccentric orbits. The GMM/Sausage component agrees well with the selected substructure stars in their chemodynamical properties, which increases our confidence in the reliability of the GMM fits.
79 ASTRONOMY AND ASTROPHYSICS↗
Enter Gaussian Mixture Modeling Extensions for Improved False Discovery Rate Estimation in GC-MS Metabolomics
Identifying small molecules (e.g., metabolites) is key towards driving scientific advancement in metabolomics, and gas chromatography–mass spectrometry (GC-MS) is an analytic method that may be applied to facilitate this process. The typical GC-MS identification workflow involves quantifying the similarity of an observed sample spectrum and other features (e.g. retention index) to that of several references, noting the compound of the best-matching reference spectrum as the identified metabolite. While a deluge of similarity metrics exists, none characterize the error rate of generated identifications, thereby presenting an unknown risk of false identification or discovery. To quantify this unknown risk, we propose a model-based framework for estimating the false discovery rate (FDR) among a set of identifications. Extending the traditional mixture modeling framework, our method incorporates both similarity score and experimental information in estimating the FDR. We apply these models to identification lists derived from across 548 samples of varying complexity and sample type (e.g., fungal species, standard mixtures, etc.), comparing their performance to that of the traditional Gaussian mixture model (GMM). Through simulation, we additionally assess the impact of reference library size on the accuracy of FDR estimates. In comparing the best performing model extensions to the GMM, our results indicate relative decreases in median absolute estimation error (MAE) ranging from 12% to 70%, based on comparisons of the median MAEs across all hit-lists. Results indicate that these relative performance improvements generally hold despite library size, however FDR estimation error typically worsens as the set of reference compounds diminishes.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗