Welcome on the homepage of the PASCAL Bootcamp 2010
Machine Learning is a multidisciplinary field which involves theoretical works linked with language theory, database theory, combinatorial optimisation, artificial intelligence, information theory but also with inferential statistics or, more recently, signal processing.
This discipline aims at extracting statistical regularities from a finite sample of observations in preparation from an inference process, where algorithmic, theoretical and practical aspects are major concerns.
Application fields of machine learning can be potentially extended to all problems where an automatic processing of data (numeric or structured) is required. It mainly includes data classification problems, document indexations, information retrieval, automatic text summarization, bio-informatics (such as protein structures prediction), automatic processing of natural language, information extraction from multimedia data, audio/video signal modelisation (source separation, sequence classification).
Most of these domains are upstream from data mining and web mining, two booming sectors of computer science with major economic issues.
Up to now, Machine Learning researchers get an inadequate academical formation. Indeed, most of european computer sciences Masters integrate only few machine learning modules, and are far from encompassing the global nature of our discipline. It is essential that PhD students (or students just about to start a PhD) get the opportunity to top up their formation with events such as this bootcamp.
This bootcamp is therefore designed for machine learning PhD students but also to PhD students from connate fields (statistics, bioinformatics, natural language processing, optimisation, signal processing...). Participants will receive a set of lessons on several major machine learning themes given by european experts from the PASCAL2 network.