Optimization and evaluation of probabilistic-logic sequence models

Publikation: KonferencebidragPaperForskning

Resumé

Analysis of biological
sequence data demands more and more sophisticated and fine-grained models,
but these in turn introduce hard computational problems.
A class of probabilistic-logic models is considered, which increases the expressibility
from HMM's and SCFG's regular and context-free languages to,
in principle, Turing complete languages.
In general, such models are computationally far to complex for direct use, so
optimization by pruning and approximation are needed.
%
The first steps are made towards a methodology for optimizing such models
by approximations
using auxiliary models for preprocessing or splitting them into submodels.
An evaluation method for approximating models is suggested
based on automatic generation of samples.
These models and evaluation processes are illustrated in the PRISM system
developed by other authors.

OriginalsprogEngelsk
Publikationsdato2008
Antal sider9
StatusUdgivet - 2008
BegivenhedStatistical and Relational Learning in Bioinformatics; Workshop of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - Antwerpen, Belgien
Varighed: 15 sep. 200819 sep. 2008

Konference

KonferenceStatistical and Relational Learning in Bioinformatics; Workshop of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
LandBelgien
ByAntwerpen
Periode15/09/200819/09/2008

Citer dette

Christiansen, H., & Lassen, O. T. (2008). Optimization and evaluation of probabilistic-logic sequence models. Afhandling præsenteret på Statistical and Relational Learning in Bioinformatics; Workshop of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Antwerpen, Belgien.
Christiansen, Henning ; Lassen, Ole Torp. / Optimization and evaluation of probabilistic-logic sequence models. Afhandling præsenteret på Statistical and Relational Learning in Bioinformatics; Workshop of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Antwerpen, Belgien.9 s.
@conference{60fde940dbcb11dd87fb000ea68e967b,
title = "Optimization and evaluation of probabilistic-logic sequence models",
abstract = "Analysis of biologicalsequence data demands more and more sophisticated and fine-grained models,but these in turn introduce hard computational problems.A class of probabilistic-logic models is considered, which increases the expressibilityfrom HMM's and SCFG's regular and context-free languages to,in principle, Turing complete languages.In general, such models are computationally far to complex for direct use, sooptimization by pruning and approximation are needed.{\%}The first steps are made towards a methodology for optimizing such modelsby approximationsusing auxiliary models for preprocessing or splitting them into submodels.An evaluation method for approximating models is suggestedbased on automatic generation of samples.These models and evaluation processes are illustrated in the PRISM systemdeveloped by other authors.",
author = "Henning Christiansen and Lassen, {Ole Torp}",
year = "2008",
language = "English",
note = "null ; Conference date: 15-09-2008 Through 19-09-2008",

}

Christiansen, H & Lassen, OT 2008, 'Optimization and evaluation of probabilistic-logic sequence models' Paper fremlagt ved Statistical and Relational Learning in Bioinformatics; Workshop of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Antwerpen, Belgien, 15/09/2008 - 19/09/2008, .

Optimization and evaluation of probabilistic-logic sequence models. / Christiansen, Henning; Lassen, Ole Torp.

2008. Afhandling præsenteret på Statistical and Relational Learning in Bioinformatics; Workshop of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Antwerpen, Belgien.

Publikation: KonferencebidragPaperForskning

TY - CONF

T1 - Optimization and evaluation of probabilistic-logic sequence models

AU - Christiansen, Henning

AU - Lassen, Ole Torp

PY - 2008

Y1 - 2008

N2 - Analysis of biologicalsequence data demands more and more sophisticated and fine-grained models,but these in turn introduce hard computational problems.A class of probabilistic-logic models is considered, which increases the expressibilityfrom HMM's and SCFG's regular and context-free languages to,in principle, Turing complete languages.In general, such models are computationally far to complex for direct use, sooptimization by pruning and approximation are needed.%The first steps are made towards a methodology for optimizing such modelsby approximationsusing auxiliary models for preprocessing or splitting them into submodels.An evaluation method for approximating models is suggestedbased on automatic generation of samples.These models and evaluation processes are illustrated in the PRISM systemdeveloped by other authors.

AB - Analysis of biologicalsequence data demands more and more sophisticated and fine-grained models,but these in turn introduce hard computational problems.A class of probabilistic-logic models is considered, which increases the expressibilityfrom HMM's and SCFG's regular and context-free languages to,in principle, Turing complete languages.In general, such models are computationally far to complex for direct use, sooptimization by pruning and approximation are needed.%The first steps are made towards a methodology for optimizing such modelsby approximationsusing auxiliary models for preprocessing or splitting them into submodels.An evaluation method for approximating models is suggestedbased on automatic generation of samples.These models and evaluation processes are illustrated in the PRISM systemdeveloped by other authors.

M3 - Paper

ER -

Christiansen H, Lassen OT. Optimization and evaluation of probabilistic-logic sequence models. 2008. Afhandling præsenteret på Statistical and Relational Learning in Bioinformatics; Workshop of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Antwerpen, Belgien.