1. 주 제 : Universal Newton Method

2. 연 사 : Prof. Yurii Nesterov (CORE/INMA, Catholic University of Louvain, Belgium)
              Author of 4 monographs and more than 90 refereed papers in the leading optimization journals. Winner of the triennial Dantzig Prize 2000 awarded by SIAM and Mathematical Programming Society for a research having a major impact on the field of Optimization. Awarded John von Neumann Theory Prize 2009 by INFORMS (International Society on Operations Research and Management Sciences). Charles Broyden prize 2010 (best paper in Optimization Methods and Software). Honorable Francqui Chair (Liege University 2011-2012). SIAM Outstanding paper award 2012. EURO Gold Medal 2016 (by Association of European Operations Research Societies).

3. 내 용 : In this talk we present a second-order method for unconstrained minimization of convex functions. It can be applied to functions with Holder continuous Hessians. Our main scheme is the Cubic Regularization of Newton Method, equipped with a special line-search procedure. We show that the global rate of convergence of this scheme depends continuously on the smoothness parameter. Thus, our method can be used even for minimizing functions with discontinuous Hessians. At the same time, the line-search procedure is very efficient: the average number of calls of oracle per iteration is equal to two. We show that for finding a point with small norm of the gradient, the Universal Newton Method must be equipped with a special termination criterion for the line-search.

4. 일 시 : 2016년 10월 14일 (금) 오후 4시 반

5. 장 소 : 39동  325호

6. 문 의 : 이경식교수 (optima@snu.ac.kr)