|
|
|
| e-Pub |
Previous | 
Home
Section: Application Domains
Genomics
Participants : Guillemette Marot, Alain Celisse.
With the use of high throughput technologies, more and more data are generated in molecular biology studies. Our developments are applied at several levels:
Research Program
Application Domains
New Software and Platforms
New Results
- An oracle inequality for Quasi-Bayesian Non-Negative Matrix Factorisation
- PAC-Bayesian Online Clustering
- Simpler PAC-Bayesian Bounds for Hostile Data
- Clustering categorical functional data: Application to medical discharge letters
- Simultaneous dimension reduction and multi-objective clustering
- Spatial Prediction of solar energy
- Multiple change-point detection
- Differential gene expression analysis
- New concentration inequalities for the leave-
p - A new notion of stability for learning algorithms
- Model for conditionally correlated categorical data
- Mixture model for mixed kind of data
- Degeneracy in multivariate Gaussian mixtures (complete data case)
- Degeneracy in multivariate Gaussian mixtures (missing data case)
- Data units selection in statistics
- Label switching in Bayesian mixture model estimation
- Trade-off computation time and accuracy
- Projection under pairwise control
- Matching of descriptors evolving over time
- Real-time audio sources classification
- Model-Based Co-clustering for Ordinal Data
- Computational and statistical trade-offs in change-point detection
- MixtComp software for full mixed data
- MASSICCC platform for SaaS software availability
- CoModes package for correlated categorical variables
- MixCluster package for correlated mixed variables
Bilateral Contracts and Grants with Industry
Partnerships and Cooperations
Bibliography