Capturing the dynamics of the phase transition of skyrmions with a nonstationary machine learning approach
The authors propose here a nonstationary machine learning approach specifically designed for analyzing dynamic phase transitions. By employing a subsampling technique in the short-time regime, this method overcomes the difficulty of critical slowing down. It demonstrates high effectiveness and generalization in determining critical points, and static and dynamic critical exponents in the depinning transition of skyrmions.
Authors
- Bo Zheng (ORCID: https://orcid.org/0000-0001-6080-5503)
- Lei‐Lei Nian (ORCID: https://orcid.org/0000-0002-3382-1003)
- Nengji Zhou (ORCID: https://orcid.org/0000-0002-7530-8312)
- Shiqian Hu (ORCID: https://orcid.org/0000-0002-6667-4476)
- Long Xiong (ORCID: https://orcid.org/0000-0002-8702-1976)
Institutions
- Hangzhou Normal University (CN)
- Yunnan University (CN)
- Collaborative Innovation Center of Advanced Microstructures (CN)
- Zhejiang University (CN)
Publication Details
- Journal
- Physical review. B./Physical review. B
- Published
- 2025-05-14
- DOI
- https://doi.org/10.1103/physrevb.111.184415
- Citations
- 2
- Primary Topic
- Theoretical and Computational Physics
- Type
- article
- Field-Weighted Citation Impact
- 2.52
Funders
- National Natural Science Foundation of China