Modelling Class Noise with Symmetric and Asymmetric Distributions
نویسندگان
چکیده
In classification problem, we assume that the samples around the class boundary are more likely to be incorrectly annotated than others, and propose boundaryconditional class noise (BCN). Based on the BCN assumption, we use unnormalized Gaussian and Laplace distributions to directly model how class noise is generated, in symmetric and asymmetric cases. In addition, we demonstrate that Logistic regression and Probit regression can also be reinterpreted from this class noise perspective, and compare them with the proposed models. The empirical study shows that, the proposed asymmetric models overall outperform the benchmark linear models, and the asymmetric Laplace-noise model achieves the best performance among all.
منابع مشابه
Characterization of Wedge Factors and Dose Distributions in Radiotherapy with Symmetric and Asymmetric Physical Wedged Beams of 6 MV Photon Beam
Introduction: Physical wedge by modify photon beam shape and intensity has been utilized in radiotherapy to obtain uniformly dose distribution in tumor site with reduced hot spots. Calculation of dosimetric parameters for both symmetric and asymmetric wedged fields is proved necessary during linear accelerator (Linac) commissioning. The present study aimed to achieve o...
متن کاملUsing Weighted Distributions for Modeling Skewed, Multimodal and Truncated Data
When the observations reflect a multimodal, asymmetric or truncated construction or a combination of them, using usual unimodal and symmetric distributions leads to misleading results. Therefore, distributions with ability of modeling skewness, multimodality and truncation have been in the core of interest in statistical literature, always. There are different methods to contract ...
متن کاملHessian Stochastic Ordering in the Family of multivariate Generalized Hyperbolic Distributions and its Applications
In this paper, random vectors following the multivariate generalized hyperbolic (GH) distribution are compared using the hessian stochastic order. This family includes the classes of symmetric and asymmetric distributions by which different behaviors of kurtosis in skewed and heavy tail data can be captured. By considering some closed convex cones and their duals, we derive some necessary and s...
متن کاملThe Family of Scale-Mixture of Skew-Normal Distributions and Its Application in Bayesian Nonlinear Regression Models
In previous studies on fitting non-linear regression models with the symmetric structure the normality is usually assumed in the analysis of data. This choice may be inappropriate when the distribution of residual terms is asymmetric. Recently, the family of scale-mixture of skew-normal distributions is the main concern of many researchers. This family includes several skewed and heavy-tailed d...
متن کاملModeling Stock Return Volatility Using Symmetric and Asymmetric Nonlinear State Space Models: Case of Tehran Stock Market
Volatility is a measure of uncertainty that plays a central role in financial theory, risk management, and pricing authority. Turbulence is the conditional variance of changes in asset prices that is not directly observable and is considered a hidden variable that is indirectly calculated using some approximations. To do this, two general approaches are presented in the literature of financial ...
متن کامل