Speaker
Michael Biehl
(Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen)
Description
A brief introduction to the distance based analysis and
classification of multi-dimensional data is given. The popular
Learning Vector Quantization (LVQ) will serve as the main
example in this talk.
A key issue in LVQ and many related methods is the choice of
a suitable similarity or distance measure. So-called relevance
learning schemes employ parameterized distance measures
which are optimized in a data-driven training process. The
recently introduced Matrix Relevance LVQ, based on
generalized Euclidean distances, will be discussed in greater
detail.
Matrix Relevance Learning and Divergence based LVQ are
illustrated in terms of a number of real world applications
from the biomedical domain.
Examples include the diagnosis of Acute Myeloid Leukaemia
based on flow cytometry and the detection of malignancy in
adrenal tumors from steroid metabolomics data.
Application examples illustrate how relevance learning can be
used for efficient multivariate analysis and classification, for
the identification of most relevant biomarkers, and for the
visualization of complex, labelled data sets.