Staged Mixture Modelling and Boosting

Staged Mixture Modeling and Boosting

In this paper, we introduce and evaluate a data-driven staged mixture modeling technique for building density, regression, and classi cation models. Our basic approach is to sequentially add components to a nite mixture model using the structural expectation maximization (SEM) algorithm. We show that our technique is qualitatively similar to boosting. This correspondence is a natural byproduct ...

SPORE: Staged Probabilistic Regression for Hand Orientation Inference

Learning the global hand orientation from 2D monocular images is a challenging task, as the projected hand shape is affected by a number of variations. These include inter-person hand shape and size variations, intra-person pose and style variations and self-occlusion due to varying hand orientation. Given a hand orientation dataset containing these variations, a single regressor proves to be l...

Boosting Mixture Models for Semi-supervised Learning

Skew Laplace Finite Mixture Modelling

‎This paper presents a new mixture model via considering the univariate skew Laplace distribution‎. ‎The new model can handle both heavy tails and skewness and is multimodal‎. ‎Describing some properties of the proposed model‎, ‎we present a feasible EM algorithm for iteratively‎ ‎computing maximum likelihood estimates‎. ‎We also derive the observ...

Boosting GMM and Its Two Applications

Boosting is an effecient method to improve the classification performance. Recent theoretical work has shown that the boosting technique can be viewed as a gradient descent search for a good fit in function space. Several authors have applied such viewpoint to solve the density estimation problems. In this paper we generalize such framework to a specific density model – Gaussian Mixture Model (...