Abstract
We investigate the application of the recent cluster-based acceleration methods of Wolff and of Kandel et al. to the problem of simulating spin glasses. We find the techniques offer no improvement as the clusters generated by these algorithms are infinitely large or interact infinitely strongly, respectively. We comment on the reasons for this failure.
| Original language | American English |
|---|---|
| Pages (from-to) | 4778-4780 |
| Journal | Physical Review B (Condensed Matter and Materials Physics) |
| Volume | 41 |
| Issue number | 7 |
| State | Published - 1990 |