✦ LIBER ✦

Model parameter estimation for mixture density polynomial segment models

✍ Scribed by T. Fukada; K.K. Paliwal; Y. Sagisaka

Publisher: Elsevier Science
Year: 1998
Tongue: English
Weight: 318 KB
Volume: 12
Category: Article
ISSN: 0885-2308
DOI: 10.1006/csla.1998.0047

No coin nor oath required. For personal study only.

✦ Synopsis

In this paper, we propose parameter estimation techniques for mixture density polynomial segment models (MDPSMs) where their trajectories are specified with an arbitrary regression order. MDPSM parameters can be trained in one of three different ways: (1) segment clustering; (2) expectation maximization (EM) training of mean trajectories; and (3) EM training of mean and variance trajectories. These parameter estimation methods were evaluated in TIMIT vowel classification experiments. The experimental results showed that modelling both the mean and variance trajectories is consistently superior to modelling only the mean trajectory. We also found that modelling both trajectories results in significant improvements over the conventional HMM.

📜 SIMILAR VOLUMES

Optimal designs for estimating the param

Optimal designs for estimating the parameters in weighted power-mean-mixture models

✍ R. L. J. Coetzer; W. W. Focke 📂 Article 📅 2009 🏛 John Wiley and Sons 🌐 English ⚖ 273 KB 👁 2 views

Estimating biokinetic parameters for bio

Estimating biokinetic parameters for biofilm models

📂 Article 📅 2008 🏛 John Wiley and Sons 🌐 English ⚖ 39 KB 👁 2 views

Biomolecular motors, including the linear motor kinesin, have been widely applied as actuators in a variety of nanoengineered systems. A significant challenge within this context involves the ability to start and stop the motors, and logically control their function within hybrid systems. While natu

Parameter Estimation for the Compartment

Parameter Estimation for the Compartmental Model

✍ Prof. P. Sen; Prof. R. C. Srivastava 📂 Article 📅 2007 🏛 John Wiley and Sons 🌐 English ⚖ 626 KB

An estimation procedure is obtained for a atochaatic compartmental model. Compsrtmental analysis mumes that a system may be divided into homogeneous componente, or compertments. The main theory for the compartmental syatem was studied by MATIS and €LETLEY (1971) with a discrete population in a stea

Parameter estimation for coupled multiva

Parameter estimation for coupled multivariable error models

✍ Gang Tao; Yi Ling 📂 Article 📅 1999 🏛 John Wiley and Sons 🌐 English ⚖ 161 KB 👁 3 views

An adaptive algorithm is developed for estimating the parameters of a linear multivariable error model with coupled dynamics, using estimation errors for coupling inputs. Coupled dynamics in an error equation lead to a new type of error models which have different regressors but the same parameter e

Parameter estimation for the cusp catast

Parameter estimation for the cusp catastrophe model

✍ Loren Cobb 📂 Article 📅 1981 🏛 John Wiley and Sons 🌐 English ⚖ 302 KB 👁 2 views

COMBINING INTERNAL AND EXTERNAL ROBOT MO

COMBINING INTERNAL AND EXTERNAL ROBOT MODELS FOR IMPROVED MODEL PARAMETER ESTIMATION

✍ X. CHENUT; J.C. SAMIN; J. SWEVERS; C. GANSEMAN 📂 Article 📅 2000 🏛 Elsevier Science 🌐 English ⚖ 182 KB

Experimental robot identi"cation techniques can principally be divided into two categories, based on the type of models they use : internal or external. Internal models relate the joint torques or forces and the motion of the robot; external models relate the reaction forces and torques on the bedpl