On the way to universal quantum computing, quantum simulators, literally "analog" quantum computers, have already been a huge success and are fulfilling Feynman's original 1982 vision of quantum computing. Each such simulator requires a dedicated experimental platform, and costs on the order of several million dollars to build. Such experiments have many interacting parts often requiring a complex rearrangement and months of work in order to perform a specified quantum computation. A widely accessible and easy to use software tool to shortcut design considerations for quantum simulator experimentalists is much needed. This project will create such a tool. Initially, the project researchers will design and release on SourceForge an open source software package centered around 1D matrix product state (MPS) and matrix product density operator (MPDO) methods, for both closed and open quantum systems, which any experimentalist can download and easily use locally to design and benchmark their quantum simulator architecture of choice. The software elements will include (i) prebuilt generalized Ising, Hubbard, and other models, (ii) multi-legged ladders to extend into quasi-1D, (iii) different time propagation methods for short and long-range interactions, and (iv) supplemental exact diagonalization and quantum trajectory methods. Secondly, the project members will create an even simpler graphical version via a web interface in collaboration with the Science Gateways Community Institute which allows any experimentalist to run quick simulations and tests on local dedicated high-performance computing resources at the Colorado School of Mines in a secure and user-friendly format. Finally, the project team will develop a key new software element in terms of a series of increasingly accurate discretization schemes to model continuum quantum simulators and mesoscopic limits, as well as control systematic error in experiments.
MPS/MPDO methods enable the treatment of outstanding quantum problems for design of new materials dubbed synthetic quantum matter, push the boundaries of quantum mechanics into strongly correlated physics, where particles and even quasiparticles lose meaning, and allow exploration of totally new realms of entangled dynamics. The software package proposed here will make these extraordinary capacities accessible to the over 150 quantum simulator experimental groups, as well as the many theoretical/computational groups investigating entangled quantum dynamics, thus greatly speeding up exploration of quantum simulator physics. The software package will provide a platform for exploring far-from-equilibrium open quantum system dynamical models, an important topic in the theoretical and computational physics community at present due to the new and expanded capacity offered by quantum simulators; and it will provide an educational venue for new graduate students entering research groups and faculty developing courses in computational physics and related areas. Broader impact activities include integration of computation across the Colorado School of Mines undergraduate physics curriculum as a model for departments across the country to achieve this AAPT/APS goal, e.g. redesigning the mathematical physics course to cover equal parts analytical and computational methods, including specific research skills for summer REUs and internships. Finally, graduate students will be trained in critical analysis for scientific problem solving and rigorous numerical techniques for high-performance computing, including open source science via a science gateway, an approach key to success in a number of arenas in society, from the materials genome initiative to the space program.
This project is supported by the Office of Advanced Cyberinfrastructure in the Directorate for Computer & Information Science and Engineering and the Physics Division, Materials Research Division, and Office of Multidisciplinary Activities in the Directorate of Mathematical and Physical Sciences.
