Detecting Label Noise in Kepler Confirmed Planet Catalog using Machine Learning
Explore the source record for details and available documents.
Exoplanet↗
SEARCH · Engineering Papers
Results for “label noise”
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.
Small Hot Jet Acoustic Rig Validation
The Small Hot Jet Acoustic Rig (SHJAR), located in the Aeroacoustic Propulsion Laboratory (AAPL) at the NASA Glenn Research Center in Cleveland, Ohio, was commissioned in 2001 to test jet noise reduction concepts at low technology readiness levels (TRL 1-3) and develop advanced measurement techniques. The first series of tests on the SHJAR were designed to prove its capabilities and establish the quality of the jet noise data produced. Towards this goal, a methodology was employed dividing all noise sources into three categories: background noise, jet noise, and rig noise. Background noise was directly measured. Jet noise and rig noise were separated by using the distance and velocity scaling properties of jet noise. Effectively, any noise source that did not follow these rules of jet noise was labeled as rig noise. This method led to the identification of a high frequency noise source related to the Reynolds number. Experiments using boundary layer treatment and hot wire probes documented this noise source and its removal, allowing clean testing of low Reynolds number jets. Other tests performed characterized the amplitude and frequency of the valve noise, confirmed the location of the acoustic far field, and documented the background noise levels under several conditions. Finally, a full set of baseline data was acquired. This paper contains the methodology and test results used to verify the quality of the SHJAR rig.
Brown, Cliff↗
AveBoost2: Boosting for Noisy Data
AdaBoost is a well-known ensemble learning algorithm that constructs its constituent or base models in sequence. A key step in AdaBoost is constructing a distribution over the training examples to create each base model. This distribution, represented as a vector, is constructed to be orthogonal to the vector of mistakes made by the pre- vious base model in the sequence. The idea is to make the next base model's errors uncorrelated with those of the previous model. In previous work, we developed an algorithm, AveBoost, that constructed distributions orthogonal to the mistake vectors of all the previous models, and then averaged them to create the next base model s distribution. Our experiments demonstrated the superior accuracy of our approach. In this paper, we slightly revise our algorithm to allow us to obtain non-trivial theoretical results: bounds on the training error and generalization error (difference between training and test error). Our averaging process has a regularizing effect which, as expected, leads us to a worse training error bound for our algorithm than for AdaBoost but a superior generalization error bound. For this paper, we experimented with the data that we used in both as originally supplied and with added label noise-a small fraction of the data has its original label changed. Noisy data are notoriously difficult for AdaBoost to learn. Our algorithm's performance improvement over AdaBoost is even greater on the noisy data than the original data.
Oza, Nikunj C.↗
Evidence of hyperbolic cosmic dust particles.
Cosmic dust data from the helicentric Pioneers 8 and 9 have been gathered for more than 7 years. A review and detailed study of these data are given and show that events which were previously labeled solar disturbance events and assumed to be noise generated by solar effects are, logically, true cosmic dust impact events. They are shown as an extension of the range of particle parameters exhibited by the time-of-flight measurements. The effects of accepting the sun-oriented events as authentic impact events are discussed.
Berg, O. E.↗
Post-embedding tem signal-to-noise ratio of S-100
We assessed the reactivity of purified S-100 antiserum in immuno-electron microscopy by counting the number of gold particles per microns 2 over inner ear tissues embedded in different media. Sections containing predominantly Schwann's cell cytoplasm and nucleus, afferent fiber axoplasm and myelin sheath of chick cochleae were reacted with anti-S-100 IgG, an antibody to a calcium binding protein of neuronal tissues, then labeled with anti-IgG-gold conjugate. This investigation was conducted because previously published procedures, unmodified, did not yield acceptable results. Preparation of all specimens was identical. Only the medium (PolyBed 812, Araldite or Spurr epoxies; and LR White, LR Gold or Lowicryl plastics) was changed. The medium was made the changing variable because antigens available in post-embedding immuno-electron microscopy are decreased by heat, either used and/or released during polymerization of the embedding medium. The results indicate that: (a) none of the embedding media above provided optimal signal-to-noise ratio for all parts of the nerve stained in the same section; (b) aggregation of gold particles over cells was highest in embedding media with high background labeling over areas devoid of tissue (noise); (c) aggregation occurred randomly throughout both cellular and acellular regions; and (d) particles aggregated less and were distributed more evenly in tissues from media yielding good ultrastructural integrity.
Non-NASA Center↗
Computerized aerodynamic design of a transonically 'quiet' blade
The high noise levels produced by helicopters are major sources of concern. There are many sources of the noise, but during high-speed forward flight, impulsive noise dominates the noise spectrum. The cause of the high-speed impulsive noise is the propagation into the far field of shock waves that form on the advancing blade. This mechanism has been labeled 'delocalization'. It has been shown, however, that by judicious design of the blade-tip planform, delocalization can be prevented. The objective of the present study is to illustrate how blade-tip configurations (both planform and airfoil shape) can be systematically varied to identify shapes that avoid delocalization and simultaneously improve aerodynamic performance. This has been done using the latest version of the ROT22 transonic, full-potential, quasi-steady, rotor flow-field code. A hypothetical modern rotor blade was postulated, and tip modifications consisting of taper, sweep, and airfoil section alterations were investigated. Planform modifications were found to be most effective in eliminating delocalization.
Tauber, M. E.↗
In-Flight Anomalies and Radiation Performance of NASA Missions - Selected Lessons Learned
This presentation addresses in-flight electronic disturbances and radiation, specifically anomaly resolution. The process for anomaly review takes into account the environment, selected parts and design, existing and/or new radiation test data, risk probability and actions to be taken. Noise spikes and the meaning of upset in a fiber optic link are also discussed.
LaBel, Kenneth A.↗
Two-Photon Flow Cytometry
Flow cytometry is a powerful technique for obtaining quantitative information from fluorescence in cells. Quantitation is achieved by assuring a high degree of uniformity in the optical excitation and detection, generally by using a highly controlled flow such as is obtained via hydrodynamic focusing. In this work, we demonstrate a two-beam, two- channel detection and two-photon excitation flow cytometry (T(sup 3)FC) system that enables multi-dye analysis to be performed very simply, with greatly relaxed requirements on the fluid flow. Two-photon excitation using a femtosecond near-infrared (NIR) laser has the advantages that it enables simultaneous excitation of multiple dyes and achieves very high signal-to-noise ratio through simplified filtering and fluorescence background reduction. By matching the excitation volume to the size of a cell, single-cell detection is ensured. Labeling of cells by targeted nanoparticles with multiple fluorophores enables normalization of the fluorescence signal and thus ratiometric measurements under nonuniform excitation. Quantitative size measurements can also be done even under conditions of nonuniform flow via a two-beam layout. This innovative detection scheme not only considerably simplifies the fluid flow system and the excitation and collection optics, it opens the way to quantitative cytometry in simple and compact microfluidics systems, or in vivo. Real-time detection of fluorescent microbeads in the vasculature of mouse ear demonstrates the ability to do flow cytometry in vivo. The conditions required to perform quantitative in vivo cytometry on labeled cells will be presented.
Zhog, Cheng Frank↗
Optical See-Through Head Mounted Display Direct Linear Transformation Calibration Robustness in the Presence of User Alignment Noise
Augmented Reality (AR) is a technique by which computer generated signals synthesize impressions that are made to coexist with the surrounding real world as perceived by the user. Human smell, taste, touch and hearing can all be augmented, but most commonly AR refers to the human vision being overlaid with information otherwise not readily available to the user. A correct calibration is important on an application level, ensuring that e.g. data labels are presented at correct locations, but also on a system level to enable display techniques such as stereoscopy to function properly [SOURCE]. Thus, vital to AR, calibration methodology is an important research area. While great achievements already have been made, there are some properties in current calibration methods for augmenting vision which do not translate from its traditional use in automated cameras calibration to its use with a human operator. This paper uses a Monte Carlo simulation of a standard direct linear transformation camera calibration to investigate how user introduced head orientation noise affects the parameter estimation during a calibration procedure of an optical see-through head mounted display.
Axholt, Magnus↗
Performance comparison for Barnes model 12-1000, Exotech model 100, and Ideas Inc. Biometer Mark 2
Results of tests show that all channels of all instruments, except channel 3 of the Biometer Mark 2, were stable in response to input signals were linear, and were adequately stable in response to temperature changes. The Biometer Mark 2 is labelled with an inappropriate description of the units measured and the dynamic range is a inappropriate for field measurements causing unnecessarily high fractional errors. This instrument is, therefore, quantization limited. The dynamic range and noise performance of the Model 12-1000 are appropriate for remote sensing field research. The field of view and performance of the Model 100A and the Model 12-1000 are satisfactory. The Biometer Mark 2 has not, as yet, been satisfactorily equipped with an acceptable field of view determining device. Neither the widely used aperture plate nor the 24 deg cone are acceptable.
Robinson, B.↗
Range data description based on multiple characteristics
An algorithm for describing range images based on Mean curvature (H) and Gaussian curvature (K) is presented. Range images are unique in that they directly approximate the physical surfaces of a real world 3-D scene. The curvature parameters are derived from the fundamental theorems of differential geometry and provides visible invariant pixel labels that can be used to characterize the scene. The sign of H and K can be used to classify each pixel into one of eight possible surface types. Due to the sensitivity of these parameters to noise the resulting HK-sing map does not directly identify surfaces in the range images and must be further processed. A region growing algorithm based on modeling the scene points with a Markov Random Field (MRF) of variable neighborhood size and edge models is suggested. This approach allows the integration of information from multiple characteristics in an efficient way. The performance of the proposed algorithm on a number of synthetic and real range images is discussed.
Al-Hujazi, Ezzet↗
Using Decision Trees to Detect and Isolate Simulated Leaks in the J-2X Rocket Engine
The goal of this work was to use data-driven methods to automatically detect and isolate faults in the J-2X rocket engine. It was decided to use decision trees, since they tend to be easier to interpret than other data-driven methods. The decision tree algorithm automatically "learns" a decision tree by performing a search through the space of possible decision trees to find one that fits the training data. The particular decision tree algorithm used is known as C4.5. Simulated J-2X data from a high-fidelity simulator developed at Pratt & Whitney Rocketdyne and known as the Detailed Real-Time Model (DRTM) was used to "train" and test the decision tree. Fifty-six DRTM simulations were performed for this purpose, with different leak sizes, different leak locations, and different times of leak onset. To make the simulations as realistic as possible, they included simulated sensor noise, and included a gradual degradation in both fuel and oxidizer turbine efficiency. A decision tree was trained using 11 of these simulations, and tested using the remaining 45 simulations. In the training phase, the C4.5 algorithm was provided with labeled examples of data from nominal operation and data including leaks in each leak location. From the data, it "learned" a decision tree that can classify unseen data as having no leak or having a leak in one of the five leak locations. In the test phase, the decision tree produced very low false alarm rates and low missed detection rates on the unseen data. It had very good fault isolation rates for three of the five simulated leak locations, but it tended to confuse the remaining two locations, perhaps because a large leak at one of these two locations can look very similar to a small leak at the other location.
Schwabacher, Mark A.↗
From Formal Errors Towards Realistic Uncertainties
Evaluation of uncertainties of geodetic parameter estimates is the problem that is not yet solved in a satisfactory way. A direct evaluation of the uncertainties derived from a least square solution is labeled "formal" and is usually biased, sometimes up to an order of magnitude. Customary, the use of formal errors for scientific analysis is discouraged. We claim that the root of the problem is neglecting off-diagonal elements in the variance-covariance matrix of the noise in the data. A careful reconstruction of the full variance-covariance matrix, including the off-diagonal terms greatly improves realism of uncertainty estimates derived from least squares. We processed the dataset of VLBI group delays and built a prior variance-covariance of the atmosphere-driven noise based on analysis of the output of NASA high-resolution numerical weather models. We found that the uncertainties of parameter estimates derived from this least square solution that uses such variance-covariance matrices become much closer to realistic errors. We consider approaches for implementation of this method in routine data analysis of space geodesy data.
Leonid Petrov↗
The high speed civil transport: A technology challenge
It has been over 17 years since Concorde entered service with British Airways and Air France. The thirteen production aircraft are still safely operating and may extend their service life into the next century. We can appreciate the technical success Concorde has achieved. However, from the onset the aircraft was unable to achieve overall economic viability. Currently the aircraft does generate a yearly profit for British Airways which is privatized. Unacceptable airport noise and sonic boom overland levels were detrimental to the aircraft performing between many worldwide city pairs. On the horizon appears a possible successor to Concorde that will need to demonstrate environmental acceptability and economic viability shortly after the turn of the century. This vehicle has been labelled the High Speed Civil Transport (HSCT). I believe that we can overcome the environmental and economic goals based on encouraging market demand forecasts and technology improvements that are materializing in current research and development programs.
Mortlock, Alan K.↗
Metric Learning for Hyperspectral Image Segmentation
We present a metric learning approach to improve the performance of unsupervised hyperspectral image segmentation. Unsupervised spatial segmentation can assist both user visualization and automatic recognition of surface features. Analysts can use spatially-continuous segments to decrease noise levels and/or localize feature boundaries. However, existing segmentation methods use tasks-agnostic measures of similarity. Here we learn task-specific similarity measures from training data, improving segment fidelity to classes of interest. Multiclass Linear Discriminate Analysis produces a linear transform that optimally separates a labeled set of training classes. The defines a distance metric that generalized to a new scenes, enabling graph-based segmentation that emphasizes key spectral features. We describe tests based on data from the Compact Reconnaissance Imaging Spectrometer (CRISM) in which learned metrics improve segment homogeneity with respect to mineralogical classes.
Compact Reconnaissance Imaging Spectrometer (CRISM↗
Advanced Signal Processing Methods Applied to Digital Mammography
The work reported here is on the extension of the earlier proposal of the same title, August 1994-June 1996. The report for that work is also being submitted. The work reported there forms the foundation for this work from January 1997 to September 1997. After the earlier work was completed there were a few items that needed to be completed prior to submission of a new and more comprehensive proposal for further research. Those tasks have been completed and two new proposals have been submitted, one to NASA, and one to Health & Human Services WS). The main purpose of this extension was to refine some of the techniques that lead to automatic large scale evaluation of full mammograms. Progress on each of the proposed tasks follows. Task 1: A multiresolution segmentation of background from breast has been developed and tested. The method is based on the different noise characteristics of the two different fields. The breast field has more power in the lower octaves and the off-breast field behaves similar to a wideband process, where more power is in the high frequency octaves. After the two fields are separated by lowpass filtering, a region labeling routine is used to find the largest contiguous region, the breast. Task 2: A wavelet expansion that can decompose the image without zero padding has been developed. The method preserves all properties of the power-of-two wavelet transform and does not add appreciably to computation time or storage. This work is essential for analysis of the full mammogram, as opposed to selecting sections from the full mammogram. Task 3: A clustering method has been developed based on a simple counting mechanism. No ROC analysis has been performed (and was not proposed), so we cannot finally evaluate this work without further support. Task 4: Further testing of the filter reveals that different wavelet bases do yield slightly different qualitative results. We cannot provide quantitative conclusions about this for all possible bases without further support. Task 5: Better modeling does indeed make an improvement in the detection output. After the proposal ended, we came up with some new theoretical explanations that helps in understanding when the D4 filter should be better. This work is currently in the review process. Task 6: N/A. This no longer applies in view of Tasks 4-5. Task 7: Comprehensive plans for further work have been completed. These plans are the subject of two proposals, one to NASA and one to HHS. These proposals represent plans for a complete evaluation of the methods for identifying normal mammograms, augmented with significant further theoretical work.
Stauduhar, Richard P.↗
A method for the analysis of the benefits and costs for aeronautical research and technology
A relatively simple, consistent, and reasonable methodology for performing cost-benefit analyses which can be used to guide, justify, and explain investments in aeronautical research and technology is presented. The elements of this methodology (labeled ABC-ART for the Analysis of the Benefits and Costs of Aeronautical Research and Technology) include estimation of aircraft markets; manufacturer costs and return on investment versus aircraft price; airline costs and return on investment versus aircraft price and passenger yield; and potential system benefits--fuel savings, cost savings, and noise reduction. The application of this methodology is explained using the introduction of an advanced turboprop powered transport aircraft in the medium range market in 1978 as an example.
Williams, L. J.↗
Increasing Linear Dynamic Range of a CMOS Image Sensor
A generic design and a corresponding operating sequence have been developed for increasing the linear-response dynamic range of a complementary metal oxide/semiconductor (CMOS) image sensor. The design provides for linear calibrated dual-gain pixels that operate at high gain at a low signal level and at low gain at a signal level above a preset threshold. Unlike most prior designs for increasing dynamic range of an image sensor, this design does not entail any increase in noise (including fixed-pattern noise), decrease in responsivity or linearity, or degradation of photometric calibration. The figure is a simplified schematic diagram showing the circuit of one pixel and pertinent parts of its column readout circuitry. The conventional part of the pixel circuit includes a photodiode having a small capacitance, CD. The unconventional part includes an additional larger capacitance, CL, that can be connected to the photodiode via a transfer gate controlled in part by a latch. In the high-gain mode, the signal labeled TSR in the figure is held low through the latch, which also helps to adapt the gain on a pixel-by-pixel basis. Light must be coupled to the pixel through a microlens or by back illumination in order to obtain a high effective fill factor; this is necessary to ensure high quantum efficiency, a loss of which would minimize the efficacy of the dynamic- range-enhancement scheme. Once the level of illumination of the pixel exceeds the threshold, TSR is turned on, causing the transfer gate to conduct, thereby adding CL to the pixel capacitance. The added capacitance reduces the conversion gain, and increases the pixel electron-handling capacity, thereby providing an extension of the dynamic range. By use of an array of comparators also at the bottom of the column, photocharge voltages on sampling capacitors in each column are compared with a reference voltage to determine whether it is necessary to switch from the high-gain to the low-gain mode. Depending upon the built-in offset in each pixel and in each comparator, the point at which the gain change occurs will be different, adding gain-dependent fixed pattern noise in each pixel. The offset, and hence the fixed pattern noise, is eliminated by sampling the pixel readout charge four times by use of four capacitors (instead of two such capacitors as in conventional design) connected to the bottom of the column via electronic switches SHS1, SHR1, SHS2, and SHR2, respectively, corresponding to high and low values of the signals TSR and RST. The samples are combined in an appropriate fashion to cancel offset-induced errors, and provide spurious-free imaging with extended dynamic range.
Pain, Bedabrata↗