Abstract: In a variety of applications, it is very desirable to perform a robust analysis of high-dimensional data without being harmed by the presence of a possibly larger percentage of outlying measurements. Novel robust approaches for two different scenarios will be presented: (1) A regularized MWCD (minimum weighted covariance determinant) estimator, which is a highly robust joint estimator of the expectation and covariance matrix of multivariate data, and (2) training neural networks such as regularized multilayer perceptrons or radial basis function networks. Extensive numerical experiments were performed with the novel methods. The regularization allows an interpretation within the framework of Bayesian thinking.
Who: Jan Kalina
When: 1:00 p.m. Monday, December 18
Where: The session will occur physically at the Institute of Information Theory and Automation (UTIA). Depending on the number of listeners in room 25 or 45 (café). For directions to the institute, please refer to the following link: https://www.utia.cas.cz/contacts#way
Language: Czech (if you require English, please let us know in advance)
