Matrix-Product-State Open Source Code and the Many Body Physics of Ultracold Molecules

矩阵积态开源代码和超冷分子的多体物理学

• 批准号: 1207881
• 负责人: Lincoln Carr
• 金额: $ 43万
• 依托单位: Colorado School of Mines
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-09-01 至 2017-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1207881&HistoricalAwards=false
• 关键词: Matrix; Product; State; Open; Source

We will develop matrix product state (MPS) codes for entangled quantum many-body systems in one spatial dimension including long-range interactions, necessary for the long-range dipole-dipole interaction inherent to ultracold molecules. Our open source codes will be integrated into the ALPS (Algorithms and Libraries for Physics Simulations) package as part of the ALPS collaboration. As more complex molecules approach quantum degeneracy, the number and complexity of many-body models to describe them will also increase substantially. Additionally, searching for emergent phenomena such as quantum order requires exploration of vast parameter spaces and extrapolation using many different system sizes. This requires us to rethink the way in which we design and dispatch simulations and collect and interpret data. Open source tools which are to have a long term impact must be flexible to adapt to different physical degrees of freedom, different Hamiltonians, and different dynamical processes; they must be efficient to manage large parameter exploration; and they must contain powerful tools for extracting data from large simulations. Our proposed MPS algorithms enable us to meet these challenges and be prepared for the next wave of ultracold molecular physics.Our creation of open source code for matrix product state (MPS) and related methods has the potential for new insights into strongly correlated systems, a standing problem in physics on which we have been able to make only a little progress so far. Strongly correlated systems include high-temperature superconductors, an outstanding problem for energy technology as a sizeable fraction of energy resources are squandered simply in transmission lines; high-temperature materials have the potential to alleviate this loss. MPS methods allow us to explore different models of a quantum computer, including in our main physical subject of ultracold molecules, applied to quantum computing. Finally, the training of students in rigorous numerical techniques and high-performance parallel computing is key to success in a number of arenas in society, from the space program to global climate change to monitoring and managing nation-wide outbreaks of infectious diseases. This award will support such training, and will support a principal investigator involved in that effort from the undergraduate through graduate levels.

我们将开发包括长程相互作用在内的一维纠缠量子多体系统的矩阵乘积态(MPS)代码，这是超冷分子固有的长程偶极-偶极相互作用所必需的。作为阿尔卑斯合作的一部分，我们的开源代码将被集成到ALPS(物理模拟算法和库)包中。随着更复杂的分子接近量子简并，描述它们的多体模型的数量和复杂性也将大幅增加。此外，寻找像量子有序这样的紧急现象需要探索巨大的参数空间，并使用许多不同的系统大小进行外推。这要求我们重新思考我们设计和调度模拟以及收集和解释数据的方式。要想产生长期影响，开源工具必须灵活，以适应不同的物理自由度、不同的哈密顿量和不同的动力学过程；它们必须高效地管理大参数探索；它们必须包含从大型模拟中提取数据的强大工具。我们提出的MPS算法使我们能够应对这些挑战，并为下一波超冷分子物理做好准备。我们为矩阵乘积态(MPS)和相关方法创建的开放源代码有可能对强关联系统提供新的见解，这是物理学中的一个长期问题，到目前为止，我们在这方面只取得了很小的进展。强关联系统包括高温超导体，这是能源技术的一个突出问题，因为相当大一部分能源被浪费在传输线上；高温材料有可能减轻这种损失。MPS方法允许我们探索量子计算机的不同模型，包括在我们的主要物理主题超冷分子中，应用于量子计算。最后，学生在严格的数值技术和高性能并行计算方面的培训是在社会的许多领域取得成功的关键，从空间计划到全球气候变化，再到监测和管理全国范围内的传染病爆发。这一奖项将支持这样的培训，并将支持参与这一努力的首席研究员，从本科生到研究生。