In this paper, several approaches that can be used to improve biometric authentication applications are proposed. The idea is inspired by the ensemble approach, i.e., the use of several classifiers to solve a problem. Compared to using only one classifier, the ensemble of classifiers has the advantage of reducing the overall variance of the system. Instead of using multiple classifiers, we prop...

In these lecture notes we cover the more advanced variance reduction techniques of importance sampling and stratified sampling. Importance sampling is particularly suited for estimating the probabilities of rare events and, in conjunction with stratified sampling, often results in a sample variance that is several orders of magnitude smaller than the variance of the naive Monte-Carlo estimator....

Gradient Boosting and bagging applied to regressors can reduce the error due to bias and variance respectively. Alternatively, Stochastic Gradient Boosting (SGB) and Iterated Bagging (IB) attempt to simultaneously reduce the contribution of both bias and variance to error. We provide an extensive empirical analysis of these methods, along with two alternate bias-variance reduction approaches — ...

The purpose of the present work is to calculate specific absorbed fractions using variance reduction techniques and assess the effectiveness of these techniques in improving the efficiency (i.e. reducing the statistical uncertainties) of simulation results in cases where the distance between the source and the target organs is large and/or the target organ is small. The variance reduction techn...

We consider the use of two additive control variate methods to reduce the variance of performance gradient estimates in reinforcement learning problems. The first approach we consider is the baseline method, in which a function of the current state is added to the discounted value estimate. We relate the performance of these methods, which use sample paths, to the variance of estimates based on...

In this paper, we consider improving the stochastic variance reduce gradient (SVRG) method by incorporating curvature information of objective function. We propose to gradients using computationally efficient Barzilai-Borwein (BB) it into SVRG. also incorporate a BB-step size as variant. show linear convergence not only proposed but other existing SVRG variants that use second-order information...

VARIANCE-REDUCTION TECHNIQUES FOR ESTIMATING QUANTILES AND VALUE-AT-RISK by Fang Chu Quantiles, as a performance measure, arise in many practical contexts. In finance, quantiles are called values-at-risk (VARs), and they are widely used in the financial industry to measure portfolio risk. When the cumulative distribution function is unknown, the quantile can not be computed exactly and must be ...

LIMITED DISTRIBUTION NOTICE: This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and sp...

this article focuses on the estimation of population proportion when the study variable is sensitive in nature. two implicit randomized response techniques are proposed where the unrelated trait can be chosen subjectively. in addition to unbiased estimation of population proportion and variance, an empirical study is conducted to inspect the relative efficiency facet of the proposed techniques....