Long-time asymptotics for open classical systems
- Date
- Nov 10, 2011
- Time
- 2:50 PM - 3:50 PM
- Speaker
- Dr. Grigorios Pavliotis
- Affiliation
- Imperial College London
- Series
- TUD Mathematik AG Analysis & Stochastik
- Language
- en
- Main Topic
- Mathematik
- Other Topics
- Mathematik
- Description
- In this talk I will present some recent results on the long time asympotics of a small Hamiltonian system (the "distinguished particle") coupled to a heat bath which is modeled as an infinite dimensional Hamiltonian system with random initial conditions. After elimination of the heat bath variables, the dynamics of the distinguished particle can be described using a stochastic integrodifferential equation, the generalized Langevin equation (GLE). The GLE can then be approximated by a degenerate Markov process in an extended phase space. For this Markov process we prove exponentially fast convergence to equilibrium, a homogenization theorem (invariance principle), estimates on derivatives of the associated Markov semigroup and we also study several distinguished limits of physical interest. Applications of our results to MCMC (Markov Chain Monte Carlo) techniques are also discussed. Our proofs are based on a careful analysis of the spectrum of the generator of this Markov process, which is a hypoelliptic operator, and extensive use of the recently developed theory of hypocoercivity is made. This is joint work with Michela Ottobre and Karel Pravda-Starov.
- Links
Last modified: Nov 2, 2011, 11:58:45 AM
Location
TUD Willers-Bau (WIL A 124 )Zellescher Weg12-1401069Dresden
- Homepage
- https://navigator.tu-dresden.de/etplan/wil/00
Organizer
TUD MathematikWillersbau, Zellescher Weg12-1401069Dresden
- Phone
- 49-351-463 33376
- Homepage
- http://tu-dresden.de/mathematik
Legend
- Biology
- Chemistry
- Civil Eng., Architecture
- Computer Science
- Economics
- Electrical and Computer Eng.
- Environmental Sciences
- for Pupils
- Law
- Linguistics, Literature and Culture
- Materials
- Mathematics
- Mechanical Engineering
- Medicine
- Physics
- Psychology
- Society, Philosophy, Education
- Spin-off/Transfer
- Traffic
- Training
- Welcome
