Fast algorithms for self-avoiding walks

by Nathan Clisby

Institution:
Date: Mon 18th May 2009
Time: 1:00 PM
Location: Russell Love Theatre, Richard Berry Bldg, The University of Melbourne

Abstract: Abstract: In this talk I'll give an overview of some fast algorithms for self-avoiding walks, both for enumeration and Monte Carlo simulation.
For enumeration, these algorithms include the finite lattice method, the two step method, and the application of the lace expansion, while the Monte Carlo algorithms I'll discuss include the pivot and join-and-cut algorithms for simulating fixed length walks.
The power of these methods come from clever utilisation of geometrical constraints placed on self-avoiding walk configurations.
There are interesting connections between these different methods, and I'll point out some unanswered questions and potential future research directions.