Lab Home | Phone | Search
	Center for Nonlinear Studies
	Home  ▶ People   Current   Executive Committee   Postdocs   Visitors   Students   Research   Publications  ▶ Conferences   Workshops   Sponsorship  ▶ Talks   Seminars   Postdoc Seminars Archive   Quantum Lunch   Quantum Lunch Archive   P/T Colloquia   Archive   Ulam Scholar     Postdoc Nominations  ▶ Student Requests   Student Program  ▶ Visitor Requests   Description   Past Visitors  ▶ Services   General     History of CNLS     Maps, Directions   CNLS Office   T-Division   LANL				Monday, April 04, 201611:00 AM - 12:00 PMCNLS Conference Room (TA-3, Bldg 1690) Seminar Degree Sequences Decompositions and Sampling with Fast Mixing Markov Chains (Part 1) Peter L. ErdosHungarian Academy Of Sciences Part 1: The main goal of this series of talks is to introduce the audience to some current problems of network theory. In the first talk we discuss how these problems arise from practical research. We identify five main topics. Then we will study in detail several notions related to degree sequences, and discuss some classical, fundamental results. In the second talk we will discuss further some existence and listing problems. Then we will introduce sampling problems in connection with degree sequence realizations. After a short survey of samplings based on the configuration model and an importance sampling approach, we will turn to the Markov Chain Monte Carlo (MCMC) method and will discuss some known results and methods. In the last talk we discuss MCMC decompositions in general and a new decomposition technique for degree sequences, then we will apply it to construct degree sequence classes with fast mixing swap-based Markov chain-sampling processes. Host: Nathan Lemons