MODAL - 2012 - Annual activity report

MODAL

MODAL - 2012

Project-Team Modal

Members

Overall Objectives

MOdel for Data Analysis and Learning
Highlights of the Year

Scientific Foundations

Generative model design
Data visualization

Application Domains

Application Domains

Software

Two advances for the MIXMOD software
The blockcluster package
Cuvclust package
MetaMa
SMVar
Tax3 Software
aam Program
STK++
Scan3D

New Results

Model for conditionally correlated categorical data
Model-based clustering for multivariate partial ranking data
A new probability distribution for ordinal data
Clustering and variable selection in regression
Mixture of Gaussians with Missing Data
Transfer learning in model-based clustering
Gaussian Models Scale Invariant and Stable by Projection
Decorrelating variables in high dimension for linear regression
Model-based clustering for multivariate functional data
A method to combine combinatorial optimization and statistics to mine high-throughput genotyping data
Wavelet based clustering using mixed effects functional models
Comparison of normalisation procedures in RNA-sequencing before differential analysis
Change point detection algorithm
Cross validation algorithms
Stochastic Block Model
Approximations for scan statistics.

Bilateral Contracts and Grants with Industry

Arcelor-Mittal
Gene Diffusion
ASEL & CRESGE

Partnerships and Cooperations

Regional Initiatives
National Initiatives
European Initiatives
International Research Visitors

Dissemination

Scientific Animation
Teaching - Supervision - Juries

Bibliography

Major publications
Publications of the year

Previous |

Home | Next next

Section: New Results

Stochastic Block Model

Participant : Alain Célisse.

The convergence of maximum likelihood and variational estimators in a random graph model called Stochastic Block model is addressed. To the best of our knowledge, these are the first results providing consistency for maximum likelihood and variational estimators in that model. See [5] .

Previous |

Home | Next next