Descrizione:
pattern recognition and machine learning. We will first cover the
motivations, assumptions, and basic techniques used in current
practice - covering the three main families, "wrappers", "filters" and
"embedded" methods. Once we have this foundation in place, we will
discuss more advanced topics such as how these methods relate to the
multiple classifier systems literature, and how cause/effect can be
taken into account in the feature selection process. Finally, we will
conclude with current research issues for the field, including a
unifying viewpoint on the literature from an information theoretic
perspective.
Caldendario delle lezioni:
7 Giugno, ore 9-11, aula X: Motivations and Basic Methods
1: Assessment methods and Wrappers
2: Uni-variate vs Multivariate Filter Methods
8 Giugno, ore 9-11, aula X: Advanced Filters and Embedded Methods
4: Embedded Methods
9 Giugno, ore 9-11, aula Mocci: The Variance of Feature Selection
6: Why does Feature Selection Work?
10 Giugno, ore 9-11, aula X: Probabilistic Perspectives
8: A Unifying View via Conditional Likelihood
Computer Science, University of Manchester, UK. His research career
began in the Multiple Classifier Systems field, studying the
phenomenon of `diversity'. Since this time he has diverged to study
the feature selection problem in an information theoretic framework.
Applications of his work currently include bio-health informatics and
adaptive compilers.