Hamburg Multigrid Group (HMG)

At the Theoretical Physics Department of the University of Hamburg a fairly large group is investigating the use of multigrid algorithms in Lattice Gauge Theories, i.e. quantum field theories living on a lattice of discrete sites rather than in continous space-time. For the very CPU- and storage-expensive simulations of these theories, both deterministic and stochastic versions of multigrid algorithms are needed.

Currently, research of this group focuses on these topics:

Blockspin methods, Effective Field Theories: How to define blockspins for extended objects, how to compute effective field theories.
Lattice Fermions: How to incorporate multilevel strategies in simulations with lattice fermions.
Neural Multigrid Algorithms: A multilevel algorithm variant which uses neural network techniques for optimization.
Multigrid Monte Carlo: How to improve stochastic multigrid used in Monte Carlo simulations of Lattice Gauge Theories.
Object-Oriented MG programming: Development of a C++ toolit for design and testing of MG algorithms.

Here are people and publications from the Hamburg Multigrid Group.

Last updated August 8, 1993