﻿

Seminar on Computer Science and Applied Mathematics

PROGRAM

Knez Mihajlova 36
Fakultet organizacionih nauka, Univerzitet u Beogradu,
Jove Ilica 154
IEEE Chapter Computer Science (CO-16) Belgrade, Republic of Serbia

SEMINAR ZA RAČUNARSTVO I PRIMENJENU MATEMATIKU

MI SANU, Knez Mihailova 36, sala 301f

Upravni odbor Matematickog instituta SANU je na nedavnoj sednici doneo odluku da se dosadasnji Seminar za primenjenu matematiku, sada nazove Seminar za racunarstvo i primenjenu matematiku, a u cilju potenciranja znacaja racunarstva kao jedne od oblasti delatnosti Instituta. Istovremeno, Upravni odbor doneo je odluku o osnivanju Odeljenja za racunarstvo i primenjenu matematiku i vezao rad novog odeljenja za rad Seminara za racunarstvo i primenjenu matematiku.

PLAN RADA SEMINARA ZA OKTOBAR 2012. GODINE

Utorak, 02.10.2012. u 14:15, Sala 301f, MI SANU:

Panos M. Pardalos, Center for Applied Optimization (CAO), Department of Industrial and Systems Engineering, University of Florida, USA
DETECTING CRITICAL SUBSETS (NODES, EDGES, SHORTEST PATHS, OR CLIQUES) IN LARGE NETWORKS

Abstract: In network analysis, the problem of detecting subsets of elements important to the connectivity of a network (i.e., critical elements) has become a fundamental task over the last few years. Identifying the nodes, arcs, paths, clusters, cliques, etc., that are responsible for network cohesion can be crucial for studying many fundamental properties of a network. Depending on the context, finding these elements can help to analyze structural characteristics such as, attack tolerance, robustness, and vulnerability. Furthermore we can classify critical elements based on their centrality, prestige, reputation and can determine dominant clusters and partitions.

From the point of view of robustness and vulnerability analysis, evaluating how well a network will perform under certain disruptive events plays a vital role in the design and operation of such a network. To detect vulnerability issues, it is of particular importance to analyze how well connected a network will remain after a disruptive event takes place, destroying or impairing a set of its elements. The main goal is to identify the set of critical elements that must be protected or reinforced in order to mitigate the negative impact that the absence of such elements may produce in the network. Applications are typically found in homeland security, energy grid, evacuation planning, immunization strategies, financial networks, biological networks, and transportation.

From the member-classification perspective, identifying members with a high reputation and influential power within a social network could be of great importance when designing a marketing strategy. Positioning a product, spreading a rumor, or developing a campaign against drugs and alcohol abuse may have a great impact over society if the strategy is properly targeted among the most influential and recognized members of a community. The recent emergence of social networks such as Facebook, Twitter, LinkedIn, etc. provide countless applications for problems of critical-element detection.

Utorak, 09.10.2012. u 14:15h, Sala 301f, MI SANU:

!!!TERMIN IZNAJMLJEN ODELJENJU ZA MATEMATIKU!!!

Vojkan Jaksic, McGill University
ENTROPIJA I REVERZIBILNOST U KVANTNOJ STATISTICKOJ MEHANICI

Apstract: I shall discuss how the entropy production observable can be defined for any quantum (or classical) dynamical system as a derivative of the Radon-Nikodym cocycle. For the so-called open systems, which describe the interaction of several thermal reservoirs, this definition coincides with the standard thermodynamical definition in terms of the fluxes (heat, charge, mass...) across the system. After reviewing some basic properties of the entropy pro- duction observable and non-equilibrium steady states, I shall describe the large deviation theory of the entropy production observable. The main topic will be certain symmetries (Evans-Searls and Gallavotti-Cohen) of the mo- ment generating functionals which can be interpreted as an extension of the Green-Kubo linear response formula to far from equilibrium steady states. The emphasis of the talk will be on the mathematical structure of the theory. One novelty of the results is that the classical and quantum case can be treated in parallel.

Petak, 12.10.2012. u 14:15h, Sala 301f, MI SANU:

Vanredni sastanak

S. Burcu Bozkurt, Selcuk University, Konya, Turska
"ON THE NORMALIZED LAPLACIAN EIGENVALUES OF GRAPHS"

Abstract: The main part of the lecture is a presentation of a paper to be published in "Ars Combinatoria". It contains lower and upper bounds for the spectral radius. Also some results on the specral moments of the normalized Laplacian matrix will be communicated.

Utorak, 16.10.2012. u 14:15, soba 301f, MI SANU:

Jelena Milojkovic, Republicki zavod za statistiku Beograd
INTEROPERABILNOST U ELEKTRONSKOM POSLOVANJU STATISTICKIH SISTEMA

Sadrzaj: Najvazniji zadatak zvanicne statistike jeste da pruzi realnu sliku drustvenih i ekonomskih kretanja u zemlji i da obezbedi pouzdanu osnovu za analizu i donosenje odluka na raznim nivoima drustva, od drzavne uprave i drugih institucija, preko poslovnih subjekata, sve do zainteresovanih gradjana. Zahtevi koji se danas postavljaju pred statisticke sluzbe su: veca efikasnost i ucinak sa manje resursa i uporedivost statistickih podataka izmedju razlicitih statistickih sluzbi i sistema. Velika heterogenost izmedju statistickih sistema u poslovnim procesima i podacima koji su razliciti po prirodi i koriscenim IT tehnologijama, i siroka zastupljenost takozvanog stove-pipe modela statisticke proizvodnje, u kome se za svaku statisticku oblast formira posebna i nezavisna proizvodna linija, otezavaju saradnju i interoperabilnost izmedju razlicitih statistickih sluzbi i sistema i smanjuju mogucnost adekvatnog odgovora na zahteve koji se postavljaju pred njih. Model zasnovan na servisno orjentisanoj arhitekturi realizovanoj pomocu veb servisa, podrzan metapodacima i ontologijama, moze da bude resenje za interoperabilno elektronsko poslovanje statistickih sistema.

Kljucne reci: G2G integracija, interoperabilnost statistickih sistema, SOA.

Utorak, 23.10.2012. u 14:15, soba 301f, MI SANU:

Tatjana Jaksic, Matematicki institut SANU
PRIMERI PROBLEMA KOMBINATORNE OPTIMIZACIJE U KRIPTOLOGIJI

Sadrzaj: Razvoj elektronskih komunikacija doveo je do toga da se razmena informacija poverljive sadrzine odvija svakodnevno. Iz tog razloga je bitno da prilikom npr. razmene elektronske poste ili koriscenja usluga elektronskog bankarstva, ne dodje do neovlascenog pristupa podacima koji se salju. Da bi se prenos ovakvih informacija ucinio sigurnijim, one se modifikuju na takav nacin da treca strana (osoba kojoj informacije nisu namenjene) ne moze da ih protumaci u slucaju da dodje u njihov posed.

Kriptologija je (matematicka) naucna disciplina o bezbednosti informacija koja se, pored osnovnih zadataka ocuvanja tajnosti podataka (cime se bavi kriptografija), bavi i tehnikama pristupa otvorenom tekstu bez prethodnog poznavanja kljuca ili rekonstrukcije samog tajnog kljuca (cime se bavi kriptoanaliza). Da bi se unapredila bezbednost prenosa podataka, pored drugih tehnika, koriste se i metode kombinatorne optimizacije. Cilj ovog predavnja je da ilustruje kako se neki problemi kriptologije mogu modelirati kao problemi kombinatorne optimizacije. Jedna primena je u oblasti kriptoanalize sa namerom da se identifikuju i iskoriste oni kombinatorni problemi koji nisu lako resivi i tako onemoguci trecoj strani da procita origanalan sadrzaj poruke.

Utorak, 29.05.2012. u 14:15, soba 301f, MI SANU: