INRIA Research Project Proposal mistis Modelling and Inference of Complex and Structured Stochastic Systems

5 Domains of research 10 5.1 Mixture models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 5.1.1 Learning and classification techniques . . . . . . . . . . . . . . . . . . 11 5.1.2 Taking into account the curse of dimensionality. . . . . . . . . . . . 12 5.2 Markov models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 5.2.1 Triplet Markov Fields for the classification of complex structure data 13 5.2.2 Integrated Markov models . . . . . . . . . . . . . . . . . . . . . . . . 17 5.2.3 Comparing two models for clustering spatial data . . . . . . . . . . . 18 5.2.4 EM procedures using mean field-like approximations for Markov modelbased image segmentation and clustering . . . . . . . . . . . . . . . . 19 5.2.5 Convergence properties of EM-like algorithms for inference in Hidden Markov Random Fields . . . . . . . . . . . . . . . . . . . . . . . . . . 21 5.2.6 Approximations for selecting complex structure models . . . . . . . . 21 5.3 Semi and non parametric methods . . . . . . . . . . . . . . . . . . . . . . . . 23