Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Likelihood Functions”

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 253 records · Page 14

Maximum Likelihood Estimation: Some Basics

The maximum likelihood estimation is a general estimation procedure. It is often compared to estimation procedures like the ordinary least squares regression or generalized method of moments, to name a few. We discuss some basics about the maximum likelihood estimation, its advantages and disadvantages, and provide an example application to a gamma distribution function.

97 MATHEMATICS AND COMPUTING↗

Maximum likelihood estimation of difference scaling functions for suprathreshold judgments

Maximum likelihood estimation (MLE) has been used to produce perceptual scales from binary judgments of triads and quadruples. This method relies on Thurstone’s theory of a stochastic perceptual process where the perceived difference of two stimuli is the difference in their perceived strengths. It is possible that the perception of a suprathreshold difference is overestimated when adding smaller differences, a phenomenon referred to as diminishing returns. The current approach to construct a perceptual scale using MLE does not account for this phenomenon. We present a way to model the perception of differences using MLE and Thurstone’s theory, adapted to allow the possibility of diminishing returns. This method is validated using Monte Carlo simulated responses to experimental triads and can correctly model diminishing returns, the absence of diminishing returns, and the opposite of diminishing returns both in the cases when a perceptual scale is known and when the true perceived strengths of the stimuli are unknown. Additionally, this method was applied to empirical data sets to determine its feasibility in investigations of perception. Ultimately, it was found that this analysis allows for more accurate modeling of suprathreshold difference judgments, a more complete understanding of the perceptual processes underlying comparisons, and the evaluation of Thurstone’s theory of difference judgments.

97 MATHEMATICS AND COMPUTING↗

Generalized X-ray and neutron crystallographic analysis: more accurate and complete structures for biological macromolecules

X-ray and neutron crystallographic data have been combined in a joint structure-refinement procedure that has been developed using recent advances in modern computational methodologies, including cross-validated maximum-likelihood target functions with gradient-based optimization and simulated annealing. X-ray and neutron crystallographic techniques provide complementary information on the structure and function of biological macromolecules. X-ray and neutron (XN) crystallographic data have been combined in a joint structure-refinement procedure that has been developed using recent advances in modern computational methodologies, including cross-validated maximum-likelihood target functions with gradient-based optimization and simulated annealing. The XN approach for complete (including hydrogen) macromolecular structure analysis provides more accurate and complete structures, as demonstrated for diisopropyl fluorophosphatase, photoactive yellow protein and human aldose reductase. Furthermore, this method has several practical advantages, including the easier determination of the orientation of water molecules, hydroxyl groups and some amino-acid side chains.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Measurement of the top quark mass at CDF using the 'neutrino {phi} weighting' template method on a lepton plus isolated track sample

We present a measurement of the top quark mass with tt dilepton events produced in pp collisions at the Fermilab Tevatron ({radical}(s)=1.96 TeV) and collected by the CDF II detector. A sample of 328 events with a charged electron or muon and an isolated track, corresponding to an integrated luminosity of 2.9 fb{sup -1}, are selected as tt candidates. To account for the unconstrained event kinematics, we scan over the phase space of the azimuthal angles ({phi}{sub {nu}{sub 1}},{phi}{sub {nu}{sub 2}}) of neutrinos and reconstruct the top quark mass for each {phi}{sub {nu}{sub 1}}, {phi}{sub {nu}{sub 2}} pair by minimizing a {chi}{sup 2} function in the tt dilepton hypothesis. We assign {chi}{sup 2}-dependent weights to the solutions in order to build a preferred mass for each event. Preferred mass distributions (templates) are built from simulated tt and background events, and parametrized in order to provide continuous probability density functions. A likelihood fit to the mass distribution in data as a weighted sum of signal and background probability density functions gives a top quark mass of 165.5{sub -3.3}{sup +3.4}(stat){+-}3.1(syst) GeV/c{sup 2}.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Use of Geriatric Assessment Tools in Selecting Therapies in Women Aged ≥70 Years With Hormone Receptor–Positive Early-Stage Breast Cancer: Preliminary Experience With a Quality Improvement Initiative

Purpose: To develop a multidisciplinary algorithmic approach to management of women aged ≥70 years with clinically staged T1N0 hormone receptor–positive breast cancer, including geriatric assessments predicting life expectancy and the likelihood of functional decline in the near future, in the context of a program-wide quality improvement initiative, to better select patients for therapeutic interventions. Methods and Materials: Two geriatric assessment tools, the Combined Lee-Schonberg Index and the Vulnerable Elderly Scale, were introduced into our clinical workflow to predict long-term mortality and likelihood of functional decline. Scores from these tools, along with patient preferences and clinical features, were incorporated into a preoperative algorithm addressing the use of sentinel lymph node biopsy (SLNB), and a postoperative algorithm addressing the use of adjuvant radiation therapy (RT). Results: The algorithms were approved for use in August 2015. Twenty-four patients were identified by in-clinic screening and have been managed using the algorithms as a guide. Mean patient age was 80 years (range, 71-89 years). Per the preoperative algorithm, consideration of omission of SLNB was an option in 11 of 24 patients (46%), and in total 18 of 24 (75%) opted against SLNB. Per the postsurgical algorithm, consideration of omission of adjuvant RT was an option for 19 of 24 patients (79%), and in total 17 of 24 (71%) opted to forego RT. Conclusion: Incorporation of simple geriatric assessments seems to have had a marked impact on decision making regarding both surgical and adjuvant therapies for women aged ≥70 years with early-stage hormone-positive breast cancer compared with historical patterns, with ≥71% omission of both SLNB and adjuvant RT in patients managed according to an institutional quality improvement initiative.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Maximum-likelihood analysis of the COBE angular correlation function

We have used maximum-likelihood estimation to determine the quadrupole amplitude Q(sub rms-PS) and the spectral index n of the density fluctuation power spectrum at recombination from the COBE DMR data. We find a strong correlation between the two parameters of the form Q(sub rms-PS) = (15.7 +/- 2.6) exp (0.46(1 - n)) microK for fixed n. Our result is slightly smaller than and has a smaller statistical uncertainty than the 1992 estimate of Smoot et al.

Seljak, Uros↗

Identifiability of unknown noise covariance matrices for some special cases of a linear, time-invariant, discrete-time dynamic system

Two off-line schemes are proposed for the identification of unknown noise covariance matrices Q and R of a discrete-time dynamic system. The first scheme is based on a maximum a posteriori cost function utilizing smoothed state estimates, while the second is based on a maximum likelihood cost function utilizing filtered state estimates. Sensitivity of the cost functions to Q and R is analyzed for the following cases: (1) single-input single-output systems; (2) multiinput single-output systems; and (3) single-input multioutput systems with a diagonal R. Identifiability criteria are presented for the cases considered and demonstrated by examples.

Tsang, W. L.↗

Computational aspects of maximum likelihood estimation and reduction in sensitivity function calculations

This paper discusses numerical aspects of computing maximum likelihood estimates for linear dynamical systems in state-vector form. Different gradient-based nonlinear programming methods are discussed in a unified framework and their applicability to maximum likelihood estimation is examined. The problems due to singular Hessian or singular information matrix that are common in practice are discussed in detail and methods for their solution are proposed. New results on the calculation of state sensitivity functions via reduced order models are given. Several methods for speeding convergence and reducing computation time are also discussed.

Gupta, N. K.↗

ELLIPTICAL GALAXY MASSES OUT TO FIVE EFFECTIVE RADII: THE REALM OF DARK MATTER

We estimate the masses of elliptical galaxies out to five effective radii using planetary nebulae and globular clusters as tracers. A sample of 15 elliptical galaxies with a broad variation in mass is compiled from the literature. A distribution function-maximum likelihood analysis is used to estimate the overall potential slope, normalization, and velocity anisotropy of the tracers. We assume power-law profiles for the potential and tracer density and a constant velocity anisotropy. The derived potential power-law indices lie in between the isothermal and Keplerian regime and vary with mass: there is tentative evidence that the less massive galaxies have steeper potential profiles than the more massive galaxies. We use stellar mass-to-light ratios appropriate for either a Chabrier/KTG (Kroupa, Tout and Gilmore) or Salpeter initial mass function to disentangle the stellar and dark matter components. The fraction of dark matter within five effective radii increases with mass, in agreement with several other studies. We employ simple models to show that a combination of star formation efficiency and baryon extent are able to account for this trend. These models are in good agreement with both our measurements out to five effective radii and recent Sloan Lens ACS Survey measurements within one effective radii when a universal Chabrier/KTG initial mass function is adopted.

79 ASTRONOMY AND ASTROPHYSICS↗

Systematic properties of CO emission from galaxies. I - Luminosity function

A sensitive survey of normal galaxies covering a wide range of luminosities and morphological types is combined with galaxy observations in the literature to provide a sample for statistical study. The global CO emission of these galaxies is extrapolated by modeling the galaxies with an exponential radial profile. The maximum-likelihood distribution functions of CO luminosity and CO/H I flux ratio are similar in behavior. Both have long tails of bright galaxies. However, the typical galaxy has a CO luminosity of about 10 to the 6th Jy km/s Mpc-squared or less and a CO/H I ratio of the order of 10 or less. Averages of the distributions of CO luminosity and CO/H I flux ratio are higher for galaxies of Hubble type Sb-Sbc than for groups of earlier or later types. Quantitative estimates of the possible error sources in the conversion of CO luminosity to molecular mass indicates that the peaking of CO emission at intermediate types is a fairly confident result.

Verter, Frances↗

Likelihood-based docking of models into cryo-EM maps

Optimized docking of models into cryo-EM maps requires exploiting an understanding of the signal expected in the data to minimize the calculation time while maintaining sufficient signal. The likelihood-based rotation function used in crystallography can be employed to establish plausible orientations in a docking search. A phased likelihood translation function yields scores for the placement and rigid-body refinement of oriented models. Optimized strategies for choices of the resolution of data from the cryo-EM maps to use in the calculations and the size of search volumes are based on expected log-likelihood-gain scores computed in advance of the search calculation. Tests demonstrate that the new procedure is fast, robust and effective at placing models into even challenging cryo-EM maps.

59 BASIC BIOLOGICAL SCIENCES↗

System identification and model reduction using modulating function techniques

Weighted least squares (WLS) and adaptive weighted least squares (AWLS) algorithms are initiated for continuous-time system identification using Fourier type modulating function techniques. Two stochastic signal models are examined using the mean square properties of the stochastic calculus: an equation error signal model with white noise residuals, and a more realistic white measurement noise signal model. The covariance matrices in each model are shown to be banded and sparse, and a joint likelihood cost function is developed which links the real and imaginary parts of the modulated quantities. The superior performance of above algorithms is demonstrated by comparing them with the LS/MFT and popular predicting error method (PEM) through 200 Monte Carlo simulations. A model reduction problem is formulated with the AWLS/MFT algorithm, and comparisons are made via six examples with a variety of model reduction techniques, including the well-known balanced realization method. Here the AWLS/MFT algorithm manifests higher accuracy in almost all cases, and exhibits its unique flexibility and versatility. Armed with this model reduction, the AWLS/MFT algorithm is extended into MIMO transfer function system identification problems. The impact due to the discrepancy in bandwidths and gains among subsystem is explored through five examples. Finally, as a comprehensive application, the stability derivatives of the longitudinal and lateral dynamics of an F-18 aircraft are identified using physical flight data provided by NASA. A pole-constrained SIMO and MIMO AWLS/MFT algorithm is devised and analyzed. Monte Carlo simulations illustrate its high-noise rejecting properties. Utilizing the flight data, comparisons among different MFT algorithms are tabulated and the AWLS is found to be strongly favored in almost all facets.

Shen, Yan↗

Testing of hypotheses about altitude decompression sickness by statistical analyses

This communication extends a statistical analysis of forced-descent decompression sickness at altitude in exercising subjects (J Appl Physiol 1994; 76:2726-2734) with a data subset having an additional explanatory variable, rate of ascent. The original explanatory variables for risk-function analysis were environmental pressure of the altitude, duration of exposure, and duration of pure-O2 breathing before exposure; the best fit was consistent with the idea that instantaneous risk increases linearly as altitude exposure continues. Use of the new explanatory variable improved the fit of the smaller data subset, as indicated by log likelihood. Also, with ascent rate accounted for, replacement of the term for linear accrual of instantaneous risk by a term for rise and then decay made a highly significant improvement upon the original model (log likelihood increased by 37 log units). The authors conclude that a more representative data set and removal of the variability attributable to ascent rate allowed the rise-and-decay mechanism, which is expected from theory and observations, to become manifest.

NASA Center JSC↗

Training Restricted Boltzmann Machines With a D-Wave Quantum Annealer

Restricted Boltzmann Machine (RBM) is an energy-based, undirected graphical model. It is commonly used for unsupervised and supervised machine learning. Typically, RBM is trained using contrastive divergence (CD). However, training with CD is slow and does not estimate the exact gradient of the log-likelihood cost function. In this work, the model expectation of gradient learning for RBM has been calculated using a quantum annealer (D-Wave 2000Q), where obtaining samples is faster than Markov chain Monte Carlo (MCMC) used in CD. Training and classification results of RBM trained using quantum annealing are compared with the CD-based method. The performance of the two approaches is compared with respect to the classification accuracies, image reconstruction, and log-likelihood results. The classification accuracy results indicate comparable performances of the two methods. Image reconstruction and log-likelihood results show improved performance of the CD-based method. It is shown that the samples obtained from quantum annealer can be used to train an RBM on a 64-bit “bars and stripes” dataset with classification performance similar to an RBM trained with CD. Though training based on CD showed improved learning performance, training using a quantum annealer could be useful as it eliminates computationally expensive MCMC steps of CD.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

The computation of generalized cross-validation functions through householder tridiagonalization with applications to the fitting of interaction spline models

An efficient algorithm for computing the generalized cross-validation function for the general cross-validated regularization/smoothing problem is provided. This algorithm is appropriate for problems where no natural structure is available, and the regularization/smoothing problem is solved (exactly) in a reproducing kernel Hilbert space. It is particularly appropriate for certain multivariate smoothing problems with irregularly spaced data, and certain remote sensing problems, such as those that occur in meteorology, where the sensors are arranged irregularly. The algorithm is applied to the fitting of interaction spline models with irregularly spaced data and two smoothing parameters; favorable timing results are presented. The algorithm may be extended to the computation of certain generalized maximum likelihood (GML) functions. Application of the GML algorithm to a problem in numerical weather forecasting, and to a broad class of hypothesis testing problems, is noted.

Gu, Chong↗

A statistical model for interpreting computerized dynamic posturography data

Computerized dynamic posturography (CDP) is widely used for assessment of altered balance control. CDP trials are quantified using the equilibrium score (ES), which ranges from zero to 100, as a decreasing function of peak sway angle. The problem of how best to model and analyze ESs from a controlled study is considered. The ES often exhibits a skewed distribution in repeated trials, which can lead to incorrect inference when applying standard regression or analysis of variance models. Furthermore, CDP trials are terminated when a patient loses balance. In these situations, the ES is not observable, but is assigned the lowest possible score--zero. As a result, the response variable has a mixed discrete-continuous distribution, further compromising inference obtained by standard statistical methods. Here, we develop alternative methodology for analyzing ESs under a stochastic model extending the ES to a continuous latent random variable that always exists, but is unobserved in the event of a fall. Loss of balance occurs conditionally, with probability depending on the realized latent ES. After fitting the model by a form of quasi-maximum-likelihood, one may perform statistical inference to assess the effects of explanatory variables. An example is provided, using data from the NIH/NIA Baltimore Longitudinal Study on Aging.

NASA Discipline Neuroscience↗

Relationship of the time course of venous gas bubbles to altitude decompression illness

The correlation is low between the occurrence of gas bubbles in the pulmonary artery, called venous gas emboli (VGE), and subsequent decompression illness (DCI). The correlation improves when a "grade" of VGE is considered; a zero to four categorical classification based on the intensity and duration of the VGE signal from a Doppler bubble detector. Additional insight about DCI might come from an analysis of the time course of the occurrence of VGE. Using the NASA Hypobaric Decompression Sickness Databank, we compared the time course of the VGE outcome between 322 subjects who exercised and 133 Doppler technicians who did not exercise to evaluate the role of physical activity on the VGE outcome and incidence of DCI. We also compared 61 subjects with VGE and DCI with 110 subjects with VGE but without DCI to identify unique characteristics about the time course of the VGE outcome to try to discriminate between DCI and no-DCI cases. The VGE outcome as a function of time showed a characteristic short lag, rapid response, and gradual recovery phase that was related to physical activity at altitude and the presence or absence of DCI. The average time for DCI symptoms in a limb occurred just before the time of the highest fraction of VGE in the pulmonary artery. It is likely, but not certain, that an individual will report a DCI symptom if VGE are detected early in the altitude exposure, the intensity or grade of VGE rapidly increases from a limb region, and the intensity or grade of VGE remains high.

NASA Discipline Environmental Health↗

Brain segmentation and the generation of cortical surfaces

This paper describes methods for white matter segmentation in brain images and the generation of cortical surfaces from the segmentations. We have developed a system that allows a user to start with a brain volume, obtained by modalities such as MRI or cryosection, and constructs a complete digital representation of the cortical surface. The methodology consists of three basic components: local parametric modeling and Bayesian segmentation; surface generation and local quadratic coordinate fitting; and surface editing. Segmentations are computed by parametrically fitting known density functions to the histogram of the image using the expectation maximization algorithm [DLR77]. The parametric fits are obtained locally rather than globally over the whole volume to overcome local variations in gray levels. To represent the boundary of the gray and white matter we use triangulated meshes generated using isosurface generation algorithms [GH95]. A complete system of local parametric quadratic charts [JWM+95] is superimposed on the triangulated graph to facilitate smoothing and geodesic curve tracking. Algorithms for surface editing include extraction of the largest closed surface. Results for several macaque brains are presented comparing automated and hand surface generation. Copyright 1999 Academic Press.

NASA Discipline Neuroscience↗