Conjugate gradient heat bath for ill-conditioned actions
We present a method for performing sampling from a Boltzmann distribution of an ill-conditioned quadratic action. This method is based on heat-bath thermalization along a set of conjugate directions, generated via a conjugate-gradient procedure. The resulting scheme outperforms local updates for mat...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
2007
|
| _version_ | 1797102742947233792 |
|---|---|
| author | Ceriotti, M Bussi, G Parrinello, M |
| author_facet | Ceriotti, M Bussi, G Parrinello, M |
| author_sort | Ceriotti, M |
| collection | OXFORD |
| description | We present a method for performing sampling from a Boltzmann distribution of an ill-conditioned quadratic action. This method is based on heat-bath thermalization along a set of conjugate directions, generated via a conjugate-gradient procedure. The resulting scheme outperforms local updates for matrices with very high condition number, since it avoids the slowing down of modes with lower eigenvalue, and has some advantages over the global heat-bath approach, compared to which it is more stable and allows for more freedom in devising case-specific optimizations. © 2007 The American Physical Society. |
| first_indexed | 2024-03-07T06:10:07Z |
| format | Journal article |
| id | oxford-uuid:ef30a179-fd98-4ec8-8494-c375711b0d18 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T06:10:07Z |
| publishDate | 2007 |
| record_format | dspace |
| spelling | oxford-uuid:ef30a179-fd98-4ec8-8494-c375711b0d182022-03-27T11:38:28ZConjugate gradient heat bath for ill-conditioned actionsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:ef30a179-fd98-4ec8-8494-c375711b0d18EnglishSymplectic Elements at Oxford2007Ceriotti, MBussi, GParrinello, MWe present a method for performing sampling from a Boltzmann distribution of an ill-conditioned quadratic action. This method is based on heat-bath thermalization along a set of conjugate directions, generated via a conjugate-gradient procedure. The resulting scheme outperforms local updates for matrices with very high condition number, since it avoids the slowing down of modes with lower eigenvalue, and has some advantages over the global heat-bath approach, compared to which it is more stable and allows for more freedom in devising case-specific optimizations. © 2007 The American Physical Society. |
| spellingShingle | Ceriotti, M Bussi, G Parrinello, M Conjugate gradient heat bath for ill-conditioned actions |
| title | Conjugate gradient heat bath for ill-conditioned actions |
| title_full | Conjugate gradient heat bath for ill-conditioned actions |
| title_fullStr | Conjugate gradient heat bath for ill-conditioned actions |
| title_full_unstemmed | Conjugate gradient heat bath for ill-conditioned actions |
| title_short | Conjugate gradient heat bath for ill-conditioned actions |
| title_sort | conjugate gradient heat bath for ill conditioned actions |
| work_keys_str_mv | AT ceriottim conjugategradientheatbathforillconditionedactions AT bussig conjugategradientheatbathforillconditionedactions AT parrinellom conjugategradientheatbathforillconditionedactions |