Some Global Optimization problems in Data Analysis
Thursday 7 February 2013
2:00 pm - 3:00 pm
2:00 pm - 3:00 pm
| Event type | Seminar |
| Location | John Crank - Room 128 |
Speaker: Professor Emilio Carrizosa, University of Seville, Spain
Abstract
Many Statistical Data Analysis tasks lead in one way or another to Global Optimization problems, which may include continuous and integer variables and may have a very large number of variables, making exact Global Optimization impossible. Different problems recently addressed by our research group will be analyzed, namely, Sparse Principal Component Analysis, Parameter Selection in Support Vector Machines, Supervised Classification with robust criteria, or finding central points on networks. The problems will be described, formulated, and some algorithmic ideas will be discussed.
