arXiv：cs.LG· Jacob Taylor, Haining Pan, Sankar Das Sarma·· 4 小时前AI 评分25
用无监督与监督学习识别 Majorana 拓扑
Machine learning Majorana topology using unsupervised and supervised learning
AI 导读
研究将无监督与监督学习结合，证明在真实短无序纳米线的 Majorana 劈裂模拟数据中，无标注数据即可区分"拓扑"与"平庸"相，并定位两者在相关参数空间中的交叉点。该方法或可用于在实验 Majorana 纳米线中识别拓扑，论文发表于 Phys. Rev. B 114, 165418。
正文
Abstract:In unsupervised learning, the training data for deep learning does not come with any labels, thus forcing the algorithm to discover hidden patterns in the data for discerning useful information. This, in principle, could be a powerful tool in identifying topological order since topology does not always manifest in obvious physical ways (e.g., topological superconductivity) for its decisive confirmation. The problem, however, is that unsupervised learning is a difficult challenge, necessitating huge computing resources, which may not always work. In the current work, we combine unsupervised and supervised learning to establish that unlabeled (simulated) data in the Majorana splitting in realistic short disordered nanowires may enable not only a distinction between `topological' and `trivial', but also where their crossover happens in the relevant parameter space. This may be a useful tool in identifying topology in experimental Majorana nanowires.
| Comments: | 14 pages, 12 figures |
| Subjects: | Disordered Systems and Neural Networks (cond-mat.dis-nn); Mesoscale and Nanoscale Physics (cond-mat.mes-hall); Machine Learning (cs.LG) |
| Cite as: | arXiv:2512.13825 [cond-mat.dis-nn] |
| (or arXiv:2512.13825v2 [cond-mat.dis-nn] for this version) | |
| https://doi.org/10.48550/arXiv.2512.13825 arXiv-issued DOI via DataCite |
|
| Journal reference: | Phys. Rev. B 114, 165418 (2026) |
| Related DOI: | https://doi.org/10.1103/ywsw-29xc
DOI(s) linking to related resources |
Submission history
From: Haining Pan [view email]
[v1]
Mon, 15 Dec 2025 19:07:15 UTC (1,130 KB)
[v2]
Tue, 6 Oct 2026 01:42:57 UTC (1,439 KB)
来源：arXiv：cs.LG · arxiv.org