Self-learning Monte Carlo with deep neural networks
The self-learning Monte Carlo (SLMC) method is a general algorithm to speedup MC simulations. Its efficiency has been demonstrated in various systems by introducing an effective model to propose global moves in the configuration space. In this paper, we show that deep neural networks can be naturall...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
American Physical Society
2018
|
| Online Access: | http://hdl.handle.net/1721.1/116011 https://orcid.org/0000-0003-1667-8011 https://orcid.org/0000-0001-8051-7349 https://orcid.org/0000-0002-8803-1017 |
| _version_ | 1826206885404999680 |
|---|---|
| author | Shen, Huitao Liu, Junwei Fu, Liang |
| author2 | Massachusetts Institute of Technology. Department of Physics |
| author_facet | Massachusetts Institute of Technology. Department of Physics Shen, Huitao Liu, Junwei Fu, Liang |
| author_sort | Shen, Huitao |
| collection | MIT |
| description | The self-learning Monte Carlo (SLMC) method is a general algorithm to speedup MC simulations. Its efficiency has been demonstrated in various systems by introducing an effective model to propose global moves in the configuration space. In this paper, we show that deep neural networks can be naturally incorporated into SLMC, and without any prior knowledge can learn the original model accurately and efficiently. Demonstrated in quantum impurity models, we reduce the complexity for a local update from O(β²) in Hirsch-Fye algorithm to O(βlnβ), which is a significant speedup especially for systems at low temperatures. |
| first_indexed | 2024-09-23T13:39:47Z |
| format | Article |
| id | mit-1721.1/116011 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T13:39:47Z |
| publishDate | 2018 |
| publisher | American Physical Society |
| record_format | dspace |
| spelling | mit-1721.1/1160112022-09-28T15:23:48Z Self-learning Monte Carlo with deep neural networks Shen, Huitao Liu, Junwei Fu, Liang Massachusetts Institute of Technology. Department of Physics Shen, Huitao Liu, Junwei Fu, Liang The self-learning Monte Carlo (SLMC) method is a general algorithm to speedup MC simulations. Its efficiency has been demonstrated in various systems by introducing an effective model to propose global moves in the configuration space. In this paper, we show that deep neural networks can be naturally incorporated into SLMC, and without any prior knowledge can learn the original model accurately and efficiently. Demonstrated in quantum impurity models, we reduce the complexity for a local update from O(β²) in Hirsch-Fye algorithm to O(βlnβ), which is a significant speedup especially for systems at low temperatures. United States. Department of Energy. Office of Basic Energy Sciences (Award DE-SC0010526) 2018-05-31T13:39:02Z 2018-05-31T13:39:02Z 2018-05 2018-05 2018-05-29T18:00:23Z Article http://purl.org/eprint/type/JournalArticle 2469-9950 2469-9969 http://hdl.handle.net/1721.1/116011 Shen, Huitao et al. "Self-learning Monte Carlo with deep neural networks." Physical Review B 97, 20 (May 2018): 205140 © 2018 American Physical Society https://orcid.org/0000-0003-1667-8011 https://orcid.org/0000-0001-8051-7349 https://orcid.org/0000-0002-8803-1017 en http://dx.doi.org/10.1103/PhysRevB.97.205140 Physical Review B Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. American Physical Society application/pdf American Physical Society American Physical Society |
| spellingShingle | Shen, Huitao Liu, Junwei Fu, Liang Self-learning Monte Carlo with deep neural networks |
| title | Self-learning Monte Carlo with deep neural networks |
| title_full | Self-learning Monte Carlo with deep neural networks |
| title_fullStr | Self-learning Monte Carlo with deep neural networks |
| title_full_unstemmed | Self-learning Monte Carlo with deep neural networks |
| title_short | Self-learning Monte Carlo with deep neural networks |
| title_sort | self learning monte carlo with deep neural networks |
| url | http://hdl.handle.net/1721.1/116011 https://orcid.org/0000-0003-1667-8011 https://orcid.org/0000-0001-8051-7349 https://orcid.org/0000-0002-8803-1017 |
| work_keys_str_mv | AT shenhuitao selflearningmontecarlowithdeepneuralnetworks AT liujunwei selflearningmontecarlowithdeepneuralnetworks AT fuliang selflearningmontecarlowithdeepneuralnetworks |