Laboratory of Structural Data Analysis Methods in Predictive Modeling
This laboratory was established as part of a scientific research project supported with a monetary grant awarded by the Government of the Russian Federation under a grant competition designed to provide governmental support to scientific research projects implemented under the supervision of the world's leading scientists at Russian institutions of higher learning (Resolution of the RF Government No.220 of April 9, 2010).
Grant Agreement No.: 11.G34.31.0073
Name of the institution of higher learning:
Federal state budget educational institution of higher professional learning "Moscow Institute of Physics and Technology (state university)"
Fields of scientific research:
Information technologies and computer systems
In order to successfully address a wide scope of problems in natural sciences and technology, econometrics and finance, one must be able to analyze and model large amounts of data with a complex structure. Here, structure is interpreted in a broad sense of the word, including such properties as smoothness, sparseness or small-scale dimensionality, unknown correlations, hierarchical or multiple-scale interaction, and time dependence.
Key project objectives:
1. To reduce dimensionality.
2. To select and calibrate a model.
3. To design effective models.
4. To achieve classification and clusterization within environments of major dimensionality.
Anticipated scientific outcomes:
The project will help design new methods and algorithms of predictive modeling of complex systems based on the structural adaptation idea. This approach has been already successfully used in different scientific disciplines and research fields, including construction of modern aircraft and other complex technological objects, processing of medical images, pharmacokinetics, macroeconomics, etc. Identifying and researching new areas for application of this approach is an important component of the project.
Leading scientist's full name: Vladimir Grigoryevich Spokoyny
Academic degrees and titles:
Candidate of physical and mathematical sciences, PhD, Professor (Humboldt-Universität zu Berlin)
Professor (Humboldt-Universität zu Berlin), Research group leader (Weierstrass Institute, Berlin)
Areas of scientific interest:
Mathematical statistics, information technologies and computer systems