神经网络已证明是表示量子多体系统(包括费米子系统)基态的强大工具。然而,将晶格对称性高效地整合到神经表示中仍然是一个重大挑战。在这项工作中,我们提出了一个将晶格对称性嵌入费米子波函数的框架,并展示了其同时求解基态和低能激发态的能力。使用群等变神经回流变换,我们研究了偏离半填充的方格上的 t-V 模型。我们的对称感知回流显著改善了基态能量,并为高达 10 x 10 的晶格提供了准确的低能激发。我们还计算了精确的两点密度相关函数和结构因子,以识别相变和临界点。这些发现引入了一个对研究量子材料和相变至关重要的对称感知框架。
英文摘要:
Neural networks have shown to be a powerful tool to represent the ground state of quantum many-body systems, including fermionic systems. However, efficiently integrating lattice symmetries into neural representations remains a significant challenge. In this work, we introduce a framework for embedding lattice symmetries in fermionic wavefunctions and demonstrate its ability to target both ground states and low-lying excitations. Using group-equivariant neural backflow transformations, we study the t-V model on a square lattice away from half-filling. Our symmetry-aware backflow significantly improves ground-state energies and yields accurate low-energy excitations for lattices up to 10 x 10. We also compute accurate two-point density-correlation functions and the structure factor to identify phase transitions and critical points. These findings introduce a symmetry-aware framework important for studying quantum materials and phase transitions.
Many-body \textit{T}-matrix theory of a strongly interacting spin-orbit coupled Fermi gas: Momentum-resolved radio-frequency spectroscopy and fermionic pairing
Probing the homogeneous spectral function of a strongly interacting superfluid atomic Fermi gas in a trap using phase separation and momentum resolved rf spectroscopy
Neural Network-Augmented Pfaffian Wave-functions for Scalable Simulations of Interacting Fermions
Center for Computational Quantum Physics, Flatiron Institute(Flatiron研究所计算量子物理中心);Division of Chemistry and Chemical Engineering, California Institute of Technology(加州理工学院化学与化学工程系);Center for Electronic Correlations and Magnetism, University of Augsburg(奥格斯堡大学电子关联与磁学中心);Center for Computational Mathematics, Flatiron Institute(Flatiron研究所计算数学中心);Department of Physics and Astronomy, Rutgers University(罗格斯大学物理与天文学系);Collège de France(法兰西公学院);CPHT, CNRS, École Polytechnique, IP Paris(巴黎综合理工学院CNRS CPHT实验室);DQMP, Université de Genève(日内瓦大学DQMP);University of Augsburg(奥格斯堡大学);California Institute of Technology(加州理工学院);Flatiron Institute(Flatiron研究所);Rutgers University(罗格斯大学);École Polytechnique(巴黎综合理工学院);Université de Genève(日内瓦大学)
Topological surface states interacting with bulk excitations in the Kondo insulator SmB$_6$ revealed via planar tunneling spectroscopy
六硼化钐(SmB $_6$)是一种著名的近藤绝缘体,其绝缘体来自于强电子关联,由于其拓扑性质的证据越来越多,从而具有受保护的表面态,最近引起了人们的极大关注。然而,与包括Bi $_2$ Se $_3$在内的弱相关对应物不同,仍然缺乏确凿的光谱证据。在这里,我们报告的结果,揭示了详细的光谱特性的钐$_6$平面隧道。在(001)和(011)单晶表面上获得的隧穿电导分别揭示了两个和一个狄拉克锥的线性态密度。非常值得注意的是,它被发现,这些拓扑状态没有被完全保护内的散装杂化间隙。一个现象学模型的隧道过程调用的表面态与体激发(自旋激子)的相互作用,如最近的理论预测,提供了一个一致的解释所有观察到的功能。我们的光谱研究支持并解释了在这种拓扑近藤绝缘体钐$_6$的不完全保护的表面状态的建议图片。
Ab-initio variational wave functions for the time-dependent many-electron Schrödinger equation
Institute of Physics, École Polytechnique Fédérale de Lausanne (EPFL)(洛桑联邦理工学院物理研究所);Center for Quantum Science and Engineering, École Polytechnique Fédérale de Lausanne (EPFL)(洛桑联邦理工学院量子科学与工程中心)
Andreev bound state spectroscopy of a quantum-dot-based Aharonov-Bohm interferometer with superconducting terminals
Institute of Physics, Czech Academy of Sciences(捷克科学院物理研究所);Jožef Stefan Institute(约泽夫·斯特凡研究所);Faculty of Mathematics and Physics, University of Ljubljana(卢布尔雅那大学数学与物理学院)