Tuesday, 16:15 - 16:40 h, Room: H 1012


Sergey Shpirko
Primal-dual subgradient method for huge-scale conic optimization problems and its applications in structural design

Coauthor: Yurii Nesterov


For huge-scale optimization problems, we suggest a new primal-dual
subgradient method. It generates the main minimization sequence in the
dual space. At the same time, it constructs an approximate primal
solution. Our scheme is based on the recursive updating technique
suggested recently by Nesterov. It allows a logarithmic dependence of the
total cost of subgradent iteration in the number of variables.
As an application, we consider a classical problem of finding an
optimal design of mechanical structures. Such a problem can be posed
in a conic form, with high sparsity of corresponding linear operator.


Talk 3 of the invited session Tue.3.H 1012
"Large-scale structured optimization" [...]
Cluster 17
"Nonsmooth optimization" [...]


  In particular, Texas Loans Online can cater to the needs of its residents. On the global pharmaceutical market this medicine was issued in 2003 by two companies - Eli Lilly and ICOS. Initially, permission to sell Cialis was obtained in Europe, Australia, New Zealand.