Density Estimation Using Mixtures of Mixtures of Gaussians

نویسندگان

  • Wael Abd-Almageed
  • Larry S. Davis
چکیده

In this paper we present a new density estimation algorithm using mixtures of mixtures of Gaussians. The new algorithm overcomes the limitations of the popular Expectation Maximization algorithm. The paper first introduces a new model selection criterion called the Penaltyless Information Criterion, which is based on the Jensen-Shannon divergence. Mean-shift is used to automatically initialize the means and covariances of the Expectation Maximization in order to obtain better structure inference. Finally, a locally linear search is performed using the Penalty-less Information Criterion in order to infer the underlying density of the data. The validity of the algorithm is verified using real color images.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Predicting the Liquid Density of Gas Condensates and LNG Mixtures from Equations of State

In this contribution, two approaches are followed to predict the saturated liquid density of liquefied natural gas (LNG) mixtures. In one approach, 12 cubic equations of state (EoSs), comprising the popular Peng-Robinson (PR) and Redlich-Kwong-Soave (RKS), are employed to predict the saturated liquid density of 20 LNG mixtures. In the other approach, these EoS are used in conjunction with a rec...

متن کامل

Scalable Training of Mixture Models via Coresets

How can we train a statistical mixture model on a massive data set? In this paper, we show how to construct coresets for mixtures of Gaussians and natural generalizations. A coreset is a weighted subset of the data, which guarantees that models fitting the coreset will also provide a good fit for the original data set. We show that, perhaps surprisingly, Gaussian mixtures admit coresets of size...

متن کامل

Some techniques in density estimation

Density estimation is an interdisciplinary topic at the intersection of statistics, theoretical computer science and machine learning. We review some old and new techniques for bounding sample complexity of estimating densities of continuous distributions, focusing on the class of mixtures of Gaussians and its subclasses.

متن کامل

A SIMPLE MODEL FOR THE ESTIMATION OF DIELECTRIC CONSTANTS OF BINARY SOLVENT MIXTURES

A simple and reliable method for quick estimation of the dielectric constant of a binary solvent mixture is proposed. The validity of the proposed method has been tested for a broad range of binary solvent mixtures

متن کامل

Semi Empirical Calculation of Intermolecular Potentials and Transport Properties of Some Binary and Ternary Industrial Refrigerant Mixtures

In this study the intermolecular potential energies of some environment-friendly industrial HFC refrigerants were obtained through the inversion method which is based on the corresponding states principle. These potentials were later employed in calculation of transport properties (viscosity, diffusion, thermal conductivity and thermal diffusion factor) of some binary and ternary refrigerant mi...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2006