Faster universal modeling for two source classes

A. Nowbakht; F.M.J. Willems

Faster universal modeling for two source classes

A. Nowbakht, F.M.J. Willems

Signal Processing Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

46 Downloads (Pure)

Abstract

The Universal Modeling algorithms proposed in [2] for two general classes of finite-context sources are reviewed. The above methods were constructed by viewing a model structure as a partition of the context space and realizing that a partition can be reached through successive splits. Here we start by constructing recursive counting algorithms to count all models belonging to the two classes and use the algorithms to perform the Bayesian Mixture. The resulting methods lead to computationally more efficient Universal Modeling algorithms.

Original language	English
Title of host publication	23rd symposium on information theory in the Benelux
Editors	B. Macq, J.-J. Quisquater
Place of Publication	Enschede
Publisher	Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)
Pages	29-36
ISBN (Print)	90-71048-16-0
Publication status	Published - 2002
Event	2002 Symposium on Information Theory in the Benelux (SITB 2002) - Louvain-la-Neuve, Belgium Duration: 29 May 2002 → 31 May 2002 Conference number: 23

Conference

Conference	2002 Symposium on Information Theory in the Benelux (SITB 2002)
Abbreviated title	SITB 2002
Country/Territory	Belgium
City	Louvain-la-Neuve
Period	29/05/02 → 31/05/02

Access to Document

Metis149590Final published version, 649 KB

Cite this

@inproceedings{1b4ece5ab0b04519a775e3c263b15c51,

title = "Faster universal modeling for two source classes",

abstract = "The Universal Modeling algorithms proposed in [2] for two general classes of finite-context sources are reviewed. The above methods were constructed by viewing a model structure as a partition of the context space and realizing that a partition can be reached through successive splits. Here we start by constructing recursive counting algorithms to count all models belonging to the two classes and use the algorithms to perform the Bayesian Mixture. The resulting methods lead to computationally more efficient Universal Modeling algorithms.",

author = "A. Nowbakht and F.M.J. Willems",

year = "2002",

language = "English",

isbn = "90-71048-16-0",

pages = "29--36",

editor = "B. Macq and J.-J. Quisquater",

booktitle = "23rd symposium on information theory in the Benelux",

publisher = "Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)",

note = "2002 Symposium on Information Theory in the Benelux (SITB 2002), SITB 2002 ; Conference date: 29-05-2002 Through 31-05-2002",

}

Faster universal modeling for two source classes. / Nowbakht, A.; Willems, F.M.J.
23rd symposium on information theory in the Benelux. ed. / B. Macq; J.-J. Quisquater. Enschede: Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), 2002. p. 29-36.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Faster universal modeling for two source classes

AU - Nowbakht, A.

AU - Willems, F.M.J.

N1 - Conference code: 23

PY - 2002

Y1 - 2002

N2 - The Universal Modeling algorithms proposed in [2] for two general classes of finite-context sources are reviewed. The above methods were constructed by viewing a model structure as a partition of the context space and realizing that a partition can be reached through successive splits. Here we start by constructing recursive counting algorithms to count all models belonging to the two classes and use the algorithms to perform the Bayesian Mixture. The resulting methods lead to computationally more efficient Universal Modeling algorithms.

AB - The Universal Modeling algorithms proposed in [2] for two general classes of finite-context sources are reviewed. The above methods were constructed by viewing a model structure as a partition of the context space and realizing that a partition can be reached through successive splits. Here we start by constructing recursive counting algorithms to count all models belonging to the two classes and use the algorithms to perform the Bayesian Mixture. The resulting methods lead to computationally more efficient Universal Modeling algorithms.

M3 - Conference contribution

SN - 90-71048-16-0

SP - 29

EP - 36

BT - 23rd symposium on information theory in the Benelux

A2 - Macq, B.

A2 - Quisquater, J.-J.

PB - Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)

CY - Enschede

T2 - 2002 Symposium on Information Theory in the Benelux (SITB 2002)

Y2 - 29 May 2002 through 31 May 2002

ER -

Faster universal modeling for two source classes

Abstract

Conference

Access to Document

Fingerprint

Cite this