Results 1  10
of
104
When Networks Disagree: Ensemble Methods for Hybrid Neural Networks
, 1993
"... This paper presents a general theoretical framework for ensemble methods of constructing significantly improved regression estimates. Given a population of regression estimators, we construct a hybrid estimator which is as good or better in the MSE sense than any estimator in the population. We argu ..."
Abstract

Cited by 349 (3 self)
 Add to MetaCart
(Show Context)
This paper presents a general theoretical framework for ensemble methods of constructing significantly improved regression estimates. Given a population of regression estimators, we construct a hybrid estimator which is as good or better in the MSE sense than any estimator in the population. We argue that the ensemble method presented has several properties: 1) It efficiently uses all the networks of a population  none of the networks need be discarded. 2) It efficiently uses all the available data for training without overfitting. 3) It inherently performs regularization by smoothing in functional space which helps to avoid overfitting. 4) It utilizes local minima to construct improved estimates whereas other neural network algorithms are hindered by local minima. 5) It is ideally suited for parallel computation. 6) It leads to a very useful and natural measure of the number of distinct estimators in a population. 7) The optimal parameters of the ensemble estimator are given in clo...
Improving Regression Estimation: Averaging Methods for Variance Reduction with Extensions to General Convex Measure Optimization
, 1993
"... ..."
On the robust estimation of power spectra, coherences, and transfer functions
 J. Geophys. Res
, 1987
"... Robust estimation of power spectra, coherences, and transfer functions is investigated in the context of geophysical data processing. The methods described are frequencydomain extensions of current techniques from the statistical iterature and are applicable in cases where sectionaveraging methods ..."
Abstract

Cited by 44 (4 self)
 Add to MetaCart
(Show Context)
Robust estimation of power spectra, coherences, and transfer functions is investigated in the context of geophysical data processing. The methods described are frequencydomain extensions of current techniques from the statistical iterature and are applicable in cases where sectionaveraging methods would be used with data that are contaminated by local nonstationarity or isolated outliers. The paper begins with a review of robust estimation theory, emphasizing statistical principles and the maximum likelihood or Mestimators. These are combined with sectionaveraging spectral techniques to obtain robust estimates of power spectra, coherences, and transfer functions in an automatic, dataadaptive fashion. Because robust methods implicitly identify abnormal data, methods for monitoring the statistical behavior of the estimation process using quantilequantile plots are also discussed. The results are illustrated using a variety of examples from electromagnetic geophysics.
Bias and Variance of Validation Methods for Function Approximation Neural Networks Under Conditions of Sparse Data
 IEEE Transactions on Systems, Man, and Cybernetics, Part C
, 1998
"... Neural networks must be constructed and validated with strong empirical dependence, which is difficult under conditions of sparse data. This paper examines the most common methods of neural network validation along with several general validation methods from the statistical resampling literature ..."
Abstract

Cited by 20 (8 self)
 Add to MetaCart
(Show Context)
Neural networks must be constructed and validated with strong empirical dependence, which is difficult under conditions of sparse data. This paper examines the most common methods of neural network validation along with several general validation methods from the statistical resampling literature as applied to function approximation networks with small sample sizes. It is shown that an increase in computation, necessary for the statistical resampling methods, produces networks that perform better than those constructed in the traditional manner. The statistical resampling methods also result in lower variance of validation, however some of the methods are biased in estimating network error. 1. INTRODUCTION To be beneficial, system models must be validated to assure the users that the model emulates the actual system in the desired manner. This is especially true of empirical models, such as neural network and statistical models, which rely primarily on observed data rather th...
Measurement of ERP latency differences: A comparison of singleparticipant and jackknifebased scoring methods
 Psychophysiology
, 2008
"... We used computer simulations to evaluate different procedures for measuring changes in the onset latency of a representative range of eventrelated components (the auditory and visual N1, P3, N2pc, and the frequencyrelated P3 difference wave). These procedures included several techniques to determi ..."
Abstract

Cited by 19 (0 self)
 Add to MetaCart
(Show Context)
We used computer simulations to evaluate different procedures for measuring changes in the onset latency of a representative range of eventrelated components (the auditory and visual N1, P3, N2pc, and the frequencyrelated P3 difference wave). These procedures included several techniques to determine onset latencies combined with approaches using both singleparticipant average waveforms and jackknifesubsample average waveforms. In general, the jackknifebased approach combined with the relative criterion technique or combined with the fractional area
Coherence of seismic body waves from local events as measured by a small aperture array
 W 2 S ′′ ( f ) (A1) and
, 1991
"... Eight local earthquakes were recorded during the operation of a smallaperture seismic array at Pinyon Flat, California. The site was chosen for its homogeneous granitic geology and its planar topography. Amplitude spectral ratios for the same signal measured at different stations had average values ..."
Abstract

Cited by 18 (4 self)
 Add to MetaCart
Eight local earthquakes were recorded during the operation of a smallaperture seismic array at Pinyon Flat, California. The site was chosen for its homogeneous granitic geology and its planar topography. Amplitude spectral ratios for the same signal measured at different stations had average values of less than 2 and maximum values of 7. Magnitudesquared coherences were estimated for all station pairs. These estimates were highest for the P wave arrivals on the vertical component and lowest for the P wave recorded on the transverse component. At 500 m station separation the P and S waves were incoherent above 15 Hz and 10 Hz, respectively. Coherence for both the P and S waves decrease as frequency increases and as distance increases. The coherence of signals from borehole sensors located at 300 and 150 m depth displays higher average coherence than equally spaced sites located on the surface. The results here suggest hat even for sites that appear to be very similar, that is, those which are located on a planar surface within a few meters of granite bedrock, the measured seismic wavefield can be distorted substantially over scale lengths of 500 m. Coherence properties were calculated from synthetic seismograms which were computed for velocity models with exponential and self similar distribution perturbations. Standard deviations of 10 % are not sufficient for the random velocity distributions to approximate the results from the smallaperture array. 1.
On the Voronoi regions of certain lattices
 MIT LIBRARIES. DOWNLOADED ON OCTOBER 28, 2008 AT 20:11 FROM IEEE XPLORE. RESTRICTIONS APPLY. ET AL.: DESIGN OF SPHERICAL LATTICE SPACE–TIME CODES 4865
, 1984
"... The Voronoi region of a lattice Ln R is the convex polytope consisting of all points of I that are closer to the origin than to any other point of Ln. In this paper we calculate the second moments of the Voronoi regions of the lattices E6*, E7*, K12, A16 and A24. The results show that these lattic ..."
Abstract

Cited by 15 (5 self)
 Add to MetaCart
The Voronoi region of a lattice Ln R is the convex polytope consisting of all points of I that are closer to the origin than to any other point of Ln. In this paper we calculate the second moments of the Voronoi regions of the lattices E6*, E7*, K12, A16 and A24. The results show that these lattices are
A statistical perspective on algorithmic leveraging
, 2013
"... One popular method for dealing with largescale data sets is sampling. Using the empirical statistical leverage scores as an importance sampling distribution, the method of algorithmic leveraging samples and rescales data matrices to reduce the data size before performing computations on the subpr ..."
Abstract

Cited by 11 (2 self)
 Add to MetaCart
(Show Context)
One popular method for dealing with largescale data sets is sampling. Using the empirical statistical leverage scores as an importance sampling distribution, the method of algorithmic leveraging samples and rescales data matrices to reduce the data size before performing computations on the subproblem. Existing work has focused on algorithmic issues, but none of it addresses statistical aspects of this method. Here, we provide an effective framework to evaluate the statistical properties of algorithmic leveraging in the context of estimating parameters in a linear regression model. In particular, for several versions of leveragebased sampling, we derive results for the bias and variance. We show that from the statistical perspective of bias and variance, neither leveragebased sampling nor uniform sampling dominates the other. This result is particularly striking, given the wellknown result that, from the algorithmic perspective of worstcase analysis, leveragebased sampling provides uniformly superior worstcase algorithmic results, when compared with uniform sampling. Based on these theoretical results, we propose and analyze two new leveraging algorithms: one constructs a smaller leastsquares problem with “shrinked” leverage scores (SLEV), and the other solves a smaller and unweighted (or biased) leastsquares problem (LEVUNW). The empirical results indicate that our theory is a good predictor of practical performance of existing and new leveragebased algorithms and that the new algorithms achieve improved performance.
Multilabel boosting for image annotation by structural grouping sparsity
 in ACM Multimedia, 2010
"... We can obtain highdimensional heterogenous features from realworld images to describe their various aspects of visual characteristics, such as color, texture and shape etc. Different kinds of heterogenous features have different intrinsic discriminative power for image understanding. The selecti ..."
Abstract

Cited by 10 (0 self)
 Add to MetaCart
(Show Context)
We can obtain highdimensional heterogenous features from realworld images to describe their various aspects of visual characteristics, such as color, texture and shape etc. Different kinds of heterogenous features have different intrinsic discriminative power for image understanding. The selection of groups of discriminative features for certain semantics is hence crucial to make the image understanding more interpretable. This paper formulates the multilabel image annotation as a regression model with a regularized penalty. We call it Multilabel Boosting by the selection of heterogeneous features with structural Grouping Sparsity (MtBGS). MtBGS induces a (structural) sparse selection model to identify subgroups of homogenous features for predicting a certain label. Moreover, the correlations among multiple tags are utilized in MtBGS to boost the performance of multilabel annotation. Extensive experiments on public image datasets show that the proposed approach has better multilabel image annotation performance and leads to a quite interpretable model for image understanding.
Modelling Heterogeneity in Cetacean Surveys
, 2000
"... Methods for improving estimation of cetacean abundance from line transect and mark recapture surveys are proposed. Using either generalized linear or generalized additive models (GLMs or GAMs), two approaches are suggested which allow heterogeneity in the spatial distribution of cetaceans to be mode ..."
Abstract

Cited by 10 (0 self)
 Add to MetaCart
(Show Context)
Methods for improving estimation of cetacean abundance from line transect and mark recapture surveys are proposed. Using either generalized linear or generalized additive models (GLMs or GAMs), two approaches are suggested which allow heterogeneity in the spatial distribution of cetaceans to be modelled from standard line transect data. In the first approach, the transect lines are divided into smaller discrete units, and the expected number of detections in each unit is modelled using explanatory spatial covariates. In the second approach, the response is derived from the observed waiting times (or distances) between detections. Fitting this model within the usual GLM or GAM framework would require restrictive assumptions, therefore an iterative procedure is formulated which enables a realistic model to be fitted. Alternatively, it is shown how this approach can be framed as a point process model, and it is suggested how the likelihood for the observed alongtrackline distances could be maximized. The methods are illustrated using line transect data from a survey of Antarctic minke whales. A surface representing the variation in density throughout the survey region is obtained, from which abundance may be estimated by numerical integration. It is also