Statistical Machine Learning Group
Our activities correspond to the development and the application of machine learning models such as Hidden Markov Models, decision trees, etc, and more specially Multi-Layer Perceptrons. Our work concerns both theoretical and application aspects.
We work on statistical analysis of models, searching links between statistical and neural models, focusing on the generalization problem; the choice of the model, the control of the complexity, variable selection, the selection of relevant patterns in data bases. The generic problems for which we develop our models are classification, discrimination and modeling (temporal series, systems, etc). Applications
We are mainly interested on the
- Information Retrieval and Machine Learning
- User Modeling,
- Handwritting recognition
- Diagnostic of complex systems
- Semi Supervised Learning
- Social Networks Analysis
We have also worked on the speech recognition (recognition and identification), prevision and modeling of different physical phenomena, command, identification, biological sequence analysis.