Stokastisten mallien seminaari

Last modified by norrosi@helsinki_fi on 2024/02/15 12:00

Stokastisten mallien seminaari 2012-13

Helsingin yliopiston Matematiikan ja tilastotieteen laitoksella kokoontuvassa Stokastisten mallien seminaarissa käsitellään monenlaista satunnaisuuteen liittyvää matematiikkaa osanottajien kiinnostusta vastaavasti. Lukuvuoden 2012-13 aiheita ovat mm. satunnaisverkot, multifraktaalimitat, gaussiset mitat, Monte Carlo -menetelmien teoria sekä diskreettien systeemien (pseudo)satunnaisuus. Seminaarissa sekä opiskellaan eri aihepiirien perusteita että käydään läpi tuoreita papereita. Seminaarissa käsitellään myös sovelluksia tietojenkäsittelytieteen ja tietoliikenteen aloilta (esim. vertaisverkkojen hajautettuja algoritmeja).

Seminaari soveltuu matematiikan ja tietojenkäsittelytieteen opintojen loppuvaiheessa oleville, jatko-opiskelijoille ja tutkijoille, joilla on perustiedot todennäköisyyslaskennasta.

Seminaari kokoontuu torstaisin 15--17 Exactumin salissa C129 ja sitä vetävät dosentti Ilkka Norros (ilkka.norros@vtt.fi) ja professori Eero Saksman (eero.saksman@helsinki.fi). Esitelmät pidetään osallistujien mukaan englanniksi tai suomeksi.

Tänä lukuvuonna seminaari kokoontuu vetäjien ulkomaanvierailuiden vuoksi jossain määrin epäsäännöllisesti. Ensimmäinen istunto on 20.9, ja tällöin keskustellaan myös syksyn ohjelmasta.

Tulevia esitelmiä

Torstai, 30.5.2013 klo 15-17 C129

Ilkka Norros:

Ergodicity of spatial processes with deaths by random connection


Torstai, 11.4.2013 klo 15-17 C129

Ilkka Norros:

Tao's correspondence principle, II


Torstai, 4.4.2013 klo 15-17 C129

Ilkka Norros:

Tao's correspondence principle


Torstai, 21.3.2013 klo 15-17 C129

Dario Gasbarra:

Poissonization in Rice noise model and diffusion tensor imaging


Torstai, 14.3.2013 klo 15-17 C129

Nicola Rebagliati:

"Computing the Graph Edit Distance Using Evolutionary Stable Strategies, Part II"


Torstai, 28.2.2013 klo 15-17 C129

Nicola Rebagliati:

"Computing the Graph Edit Distance Using Evolutionary Stable Strategies, Part I"


Torstai, 7.2.2013 klo 15--17 C129

Ilkka Norros:

 

"On the broketower process"


Torstai, 24.1.2013 klo 15--17 C129

Harri Hakula:

 

"Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions, Part 1"


Perjantai, 30.11.2012 klo klo 12--14 B119

Miika Nikula:

"On the tail of multiplicative chaos (part II)"


Torstai, 22.11.2012 klo klo 15--17 C129

Eero Saksman:

"On the tail of multiplicative chaos (part I)"


Torstai, 04.10.2012 klo klo 15--17 C129

Harri Hakula:

"New Insights from Direct Simulation of Lorentz Lattice Gas Models"


Torstai, 20.09.2012 klo klo 15--17 C129

Ilkka Norros:

"Denisov's theorem and Gaussian queues"


Torstai, 03.05.2012 klo klo 15--17 B322

Ilkka Norros:

"Superscalable file-sharing"


Torstai, 05.05.2012 klo klo 15--17 B322

Ilkka Norros:

"Superscalable file-sharing"


Torstai, 26.04.2012 klo klo 15--17 B322

Miika Nikula:

"On normalized Mandelbrojt cascade measures"


Tiistai, 17.04.2012 klo 16--18 C123 (HUOM mmuuttunut aika ja paikka!)

Julien Barral:

"On limiting laws of supercritical cascades (part II)"


Torstai, 12.04.2012 klo 15--17 B322

Harri Hakula:

"On Lorentz Lattice Gas Models"


Torstai, 29.03.2012 klo 15--17 B322

Ilkka Norros:

"Chaos by cellular automata, part II"


Torstai, 22.03.2012 klo 15--17 B322

Ilkka Norros:

"Chaos by cellular automata"


Torstai, 15.03.2012 klo 15--17 B322

Hannu Reittu:

"On an approach to graph compression motivated by Szemerédi's regularity lemma and Rissanen’s minimum description length principle"

Abstract:

Szemerédi's regularity lemma is a key result in graph theory. It suggest, roughly speaking, that any large enough graph can be 'compressed' by partitioning its vertices into a bounded number of sets in such a way that edges between set pairs are distributed in a very uniform fashion. This could point an interesting direction in study of real life networks. Namely, searching for such a compressions of networks. A problem here is what is the exact target? We discuss one approach that is based on Rissanen's minimum description length (MDL) principle. It suggests that in such a fitting problem one should consider a trade-off between accuracy and complexity of modelling in such a way that the description length is minimized. This could result in a well-founded approach in defining graph compression and replacing heuristics like the Akaike information criterion in such modelling tasks.


Torstai, 01.03.2012 klo 15--17 B322

Patrik Noren:

"Ideals of graph homomorphisms"

Abstract:
 There are many examples of algebraic objects arising from graphs. Ideals of graph
 homomorphisms are nicely behaved examples of such objects, they unify and generalize the
 toric ideals appearing in algebraic statistics. We give some examples and basic theorems
 that can be proved about these ideals of graph homomorphisms.


Torstai, 16.02.2012 klo 15--17 B322

Alexander Engström:

"Algebraic statistics for graphical models"


Torstai, 09.02.2012 klo 15--17 B322

Ilkka Norros:

"Spatial peer-to-peer file-sharing, part III"


Torstai, 02.02.2012 klo 15--17 B322

Ilkka Norros:

"Spatial peer-to-peer file-sharing, part II"


Torstai, 26.01.2012 klo 15--17 B322

Ilkka Norros:

"Spatial peer-to-peer file-sharing, part I"


Torstai, 16.12.2011 klo 15--17 B322

Peng Mei:

"Forest Formula and its applications II"


Torstai, 1.12.2011 klo 15--17 B322

Peng Mei:

"Forest Formula and its applications I"



Torstai, 24.11.2011 klo 15--17 B322

Dario Gasbarra:

"On Kotecky-Press cluster expansion"


Torstai, 17.11.2011 klo 15--17 B322

Miika Nikula:

"Introduction to Multiplicative Cascades - Part 3"


Torstai, 3.11.2011 klo 15--17 B322

Georgiy Shevchenko:

"Some multifractional non-Gaussian processes and fields"


Torstai, 13.10.2011 klo 15--17 B322

Miika Nikula:

"Introduction to Multiplicative Cascades - Part 1"


Torstai, 6.10.2011 klo 15--17 B322

Hannu Reittu:

"Tao's probabilistic and information theoretic view on Szemerédi's regularity lemma - Part 2"


Torstai, 30.9.2011 klo 15--17 B322

Hannu Reittu:

"Tao's probabilistic view on Szemerédi's regularity lemma"


Torstai, 23.9.2011 klo 15--17 B322

Harri Varpanen:

"Juggler's exclusion process"

ABSTRACT: Juggler's exclusion process describes a system of particles on the positive integers where particles drift down to zero at unit speed. After a particle hits zero, it
 jumps into a randomly chosen unoccupied site. We show that the process is ergodic if the family of jump height distributions is uniformly integrable. In a special case where the
 particles jump according to a maximum entropy principle, the equilibrium distribution can be explicitly calculated and the process reaches its equilibrium in finite time. This is
 joint work with Lasse Leskelä from University of Jyväskylä.


Torstai, 15.9.2011 klo 15--17 B322

Ilkka Norros:

"On randomness of discrete systems, part II"


Torstai, 8.9.2011 klo 15--17 B322

Ilkka Norros:

"Diskreettien systeemien satunnaisuudesta"


Tervetuloa!

Seminar on stochastic models 2011-12

The Seminar on stochastic models of the Mathematics and statistics department of the University of Helsinki studies many kinds of mathematics related to randomness, reflecting the participants' interests. The topics of 2011-12 include at least random graphs, multifractal measures, theory of Monte Carlo methods, and (pseudo)randomness in discrete systems. Basics of different areas are studied as well as recent papers. The seminar considers also applications from computer science and telecommunications (e.g., disributed algorithms of peer-to-peer networking).

The seminar is suitable for advanced and PhD students as well as researchers, with knowledge on basic probability theory.

The seminar meets on Thursdays at 15--17 in Exactum's room B322 and it is led by docent Ilkka Norros (ilkka.norros@vtt.fi) and Prof. Eero Saksman (eero.saksman@helsinki.fi). The talks are given in English or Finnish depending on the audience.

The first meeting will take place on September 8, and the program of fall 2011 will be discussed after the talk.