Support

Support Options

Submit a Support Ticket

 

Illinois PHYS 466, Lecture 12: Random Walks

By David M. Ceperley

University of Illinois at Urbana-Champaign

View Presentation (SWF)

Licensed according to this deed.

Published on

Abstract

Random Walks

Today we will discuss Markov chains (random walks), detailed balance and transition rules.

  • These methods were introduced by Metropolis et al. in 1953
    who applied it to a hard sphere liquid.
  • It is one of the most powerful and used algorithms

Content:

  • Equation of State Calculations by Fast Computing Machines
  • Markov chain or Random Walk
  • Properties of Random Walk
  • Random Walks Example from A&T 110-123
  • What is probability of being up on the second day?
  • Metropolis algorithm
  • Replace strong “Microscopic Reversibility” criterion
  • Rejection Method
  • The “Classic” Metropolis method
  • Picture of Metropolis Rejection
  • How to sample
  • MONTE CARLO CODE
  • Overview of MCMC
  • Always measure acceptance ratio. RULE: 0.1 < a.r. < 0.9
    Adjust ratio to roughly 0.5 by varying the “step size”.
  • Variance of energy (local quantity) is not as sensitive to step size.
    MC is a robust method! You don’t need to fine tune things!
  • Optimizing the moves
  • Comparison of MC and MD: Which is better?

Credits

This presentation was breezed and uploaded by Omar Sobh

Cite this work

Researchers should cite this work as follows:

  • David M. Ceperley (2009), "Illinois PHYS 466, Lecture 12: Random Walks," http://nanohub.org/resources/6544.

    BibTex | EndNote

Tags

nanoHUB.org, a resource for nanoscience and nanotechnology, is supported by the National Science Foundation and other funding agencies. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.