Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Computer Physics Communications

PairDiag: An exact diagonalization program for solving general pairing Hamiltonians

Xiaoyu Liu, Chong Qi

Published Apr 28, 2020• 12 citations
Publisher

Abstract

We present a program for solving exactly the general pairing Hamiltonian based on diagonalization. The program generates the seniority-zero shell-model-like basis vectors via the ‘01’ inversion algorithm. The Hamiltonian matrix is constructed in this seniority-zero space. The program evaluates all non-zero elements of the Hamiltonian matrix “on the fly” using the scattering operator and a search algorithm. The matrix is diagonalized by using the iterative Lanczos algorithm. The OpenMP parallel program thus developed, PairDiag, can efficiently calculate the ground-state eigenvalue and eigenvector of the general pairing Hamiltonian for both the even-mass and the odd-mass system. The program is packaged in a Fortran module, which makes it easy to use the program to replace the BCS approximation in standard self-consistent mean field calculations. For systems with dimension around 108, the calculation can be done within hours on standard desktop computers. Program Title: PairDiag. Program Files doi: https://doi.org/10.17632/dzzspfszsh.1 Licensing provisions: CC by NC 3.0. Programming language: Fortran 95. Nature of problem: The numerically exact solution of general pairing Hamiltonian can be solved by diagonalization in configuration spaces of fixed seniority. Solution method: The program constructs the seniority-zero space by the ‘01’ inversion algorithm, and diagonalizes the general pairing Hamiltonian by the Lanczos plus QR algorithm. Restrictions: The total number of orbits involved must be less than 63, and the dimension that can be calculated is restricted by the local RAM condition.

DOI
10.1016/j.cpc.2020.107349

Related papers

PairDiag: an exact diagonalization program for solving general pairing Hamiltonians

Xiaoyu Liu, Chong Qi

We present a program for solving exactly the general pairing Hamiltonian based on diagonalization. The program generates the seniority-zero shell-model-like basis vectors via the `01' inversion algorithm. The Hamiltonian matrix is constructed in that seniority-zero space. The program evaluates all non-zero elements of the Hamiltonian matrix "on the fly" using the scattering operator and the search algorithm that act on the generated basis. The matrix is diagonalized by using the iterative Lanczos algorithm. The program thus developed, PairDiag, can calculate efficiently the ground-state eigenvalue and eigenvector of any pairing Hamiltonian. The program can be easily implemented to replace the BCS approximation in standard self-consistent mean-field calculations. The code is parallelized using OpenMP. For larger systems with dimension around 10$^{8-9}$, the calculation can be done within a day on standard desktop computers.

arXiv:2001.01978•Jan 7, 2020•0 citations

On-the-fly exact diagonalization solver for quantum electronic models

Кашин Илья Владимирович, Vladimir V. Mazurenko

We propose a distinct numerical approach to effectively solve the problem of partial diagonalization of the super-large-scale quantum electronic Hamiltonian matrices. The key ingredients of our scheme are the new method for arranging the basis vectors in the computer's RAM and the algorithm allowing not to store a matrix in RAM, but to regenerate it on-the-fly during diagonalization procedure. This scheme was implemented in the program, solving the Anderson impurity model in the framework of dynamical mean-field theory (DMFT). The DMFT equations for electronic Hamiltonian with 18 effective orbitals that corresponds to the matrix with the dimension of 2.4 * 10^9 were solved on the distributed memory computational cluster.

arXiv:1508.04895•Aug 20, 2015•0 citations

Quantum spin solver near saturation: QS3

Hiroshi Ueda, Seiji Yunoki, Tokuro Shimokawa

We develop a program package named QS3 based on the (thick-restart) Lanczos method for analyzing spin-1/2 XXZ-type quantum spin models on spatially uniform/non-uniform lattices near fully polarized states, which can be mapped to dilute hardcore Bose systems. All calculations in QS3, including eigenvalue problems, expectation values for one/two-point spin operators, and static/dynamical spin structure factors, are performed in the symmetry-adapted bases specified by the number N↓ of down spins and the wave number k associated with the translational symmetry without using the bit representation for specifying spin configurations. Because of these treatments, QS3 can support large-scale quantum systems containing more than 1000 sites with dilute N↓. We show the benchmark results of QS3 for the low-energy excitation dispersion of the isotropic Heisenberg model on the 10×10×10 cubic lattice, the static and dynamical spin structure factors of the isotropic Heisenberg model on the 10×10 square lattice, and the open-MP parallelization efficiency on the supercomputer (Ohtaka) based on AMD Epyc 7702 installed at the Institute for the Solid State Physics (ISSP). Theoretical backgrounds and the user interface of QS3 are also described. Program Title: QS3 CPC Library link to program files: https://doi.org/10.17632/yc6jytvf8v.1 Developer's repository link: https://github.com/QS-Cube/ED Licensing provisions: MIT Programming language: Fortran90 External routines/libraries: BLAS, LAPACK Nature of problem: Physical properties (such as total energy, magnetic moment, two-point spin correlation function, and dynamical structure factor) Solution method: Application software based on the full diagonalization method and the exact diagonalization method using the Lanczos and thick-restart Lanczos techniques for quantum spin S=1/2 models such as the XXZ model. Restrictions: Spin S=1/2 systems with U(1) symmetry. Unusual features: Massively large quantum spin systems with U(1) symmetry near the saturation field can be solved with or without considering translational symmetry, which is difficult to treat using standard exact diagonalization libraries with the bit representation for specifying spin configurations.

arXiv:2107.00872•Computer Physics Communications•Apr 20, 2022•3 citations

Multipair approach to pairing in finite nuclear systems

Michelangelo Sambataro

A method is described to construct the ground state of a general pairing Hamiltonian for a finite nuclear systemas a product of collective, real, distinct pairs. These are determined sequentially via an iterative variational procedure that resorts to diagonalizations of the Hamiltonian in restricted model spaces. Different applications of themethod are provided that include comparisonswith exact and projected BCS results. The quantities that are examined are correlation energies, occupation numbers and pair transfer matrix elements. The method is seen to work well leading, under some circumstances, to solutions that are basically exact.

AIP conference proceedings•Jan 1, 2012•0 citations

Exact diagonalization: the Bose–Hubbard model as an example

J. M. Zhang, R X Dong

We take the Bose-Hubbard model to illustrate exact diagonalization techniques in a pedagogical way. We follow the road of first generating all the basis vectors, then setting up the Hamiltonian matrix with respect to this basis, and finally using the Lanczos algorithm to solve low lying eigenstates and eigenvalues. Emphasis is placed on how to enumerate all the basis vectors and how to use the hashing trick to set up the Hamiltonian matrix or matrices corresponding to other quantities. Although our route is not necessarily the most efficient one in practice, the techniques and ideas introduced are quite general and may find use in many other problems.

arXiv:1102.4006•European Journal of Physics•Apr 7, 2010•144 citations

Combining symmetry breaking and restoration with configuration interaction: A highly accurate many-body scheme applied to the pairing Hamiltonian

J. Ripoche, Denis Lacroix A, D. Gambacurta +2 more

Background: Ab initio many-body methods have been developed over the past ten years to address mid-mass nuclei. In their best current level of implementation, their accuracy is of the order of a few percent error on the ground-state correlation energy. Recently implemented variants of these methods are operating a breakthrough in the description of medium-mass open-shell nuclei at a polynomial computational cost while putting state-of-the-art models of internucleon interactions to the test.Purpose: As progress in the design of internucleon interactions is made, and as questions one wishes to answer are refined in connection with increasingly available experimental data, further efforts must be made to tailor many-body methods that can reach an even higher precision for an even larger number of observable quantum states or nuclei. The objective of the present work is to contribute to such a quest by designing and testing a new many-body scheme.Methods: We formulate a truncated configuration-interaction method that consists of diagonalizing the Hamiltonian in a highly truncated subspace of the total $N$-body Hilbert space. The reduced Hilbert space is generated via the particle-number projected BCS state along with projected seniority-zero two- and four-quasiparticle excitations. Furthermore, the extent by which the underlying BCS state breaks $\text{U}(1)$ symmetry is optimized in the presence of the projected two- and four-quasiparticle excitations. This constitutes an extension of the so-called restricted variation after projection method in use within the frame of multireference energy density functional calculations. The quality of the newly designed method is tested against exact solutions of the so-called attractive pairing Hamiltonian problem.Results: By construction, the method reproduces exact results for $N=2$ and $N=4$. For $N=(8,\phantom{\rule{0.16em}{0ex}}16,\phantom{\rule{0.16em}{0ex}}20)$, the error in the ground-state correlation energy is less than (0.006%, 0.1%, 0.15%) across the entire range of internucleon coupling defining the pairing Hamiltonian and driving the normal-to-superfluid quantum phase transition. The presently proposed method offers the advantage of automatic access to the low-lying spectroscopy, which it does with high accuracy.Conclusions: The numerical cost of the newly designed variational method is polynomial (${N}^{6}$) in system size. This method achieves unprecedented accuracy for the ground-state correlation energy, effective pairing gap, and one-body entropy as well as for the excitation energy of low-lying states of the attractive pairing Hamiltonian. This constitutes a sufficiently strong motivation to envision its application to realistic nuclear Hamiltonians in view of providing a complementary, accurate, and versatile ab initio description of mid-mass open-shell nuclei in the future.

arXiv:1610.04063•Physical Review C•Jan 24, 2017•21 citations
We present a program for solving exactly the general pairing Hamiltonian based on diagonalization. The program generates the seniority-zero shell-model-like basis vectors via the ‘01’ inversion algorithm. The Hamiltonian matrix is constructed in this seniority-zero space. The program evaluates all non-zero elements of the Hamiltonian matrix “on the fly” using the scattering operator and a search algorithm. The matrix is diagonalized by using the iterative Lanczos algorithm. The OpenMP parallel program thus developed, PairDiag, can efficiently calculate the ground-state eigenvalue and eigenvector of the general pairing Hamiltonian for both the even-mass and the odd-mass system. The program is packaged in a Fortran module, which makes it easy to use the program to replace the BCS approximation in standard self-consistent mean field calculations. For systems with dimension around 108, the calculation can be done within hours on standard desktop computers. Program Title: PairDiag. Program Files doi: https://doi.org/10.17632/dzzspfszsh.1 Licensing provisions: CC by NC 3.0. Programming language: Fortran 95. Nature of problem: The numerically exact solution of general pairing Hamiltonian can be solved by diagonalization in configuration spaces of fixed seniority. Solution method: The program constructs the seniority-zero space by the ‘01’ inversion algorithm, and diagonalizes the general pairing Hamiltonian by the Lanczos plus QR algorithm. Restrictions: The total number of orbits involved must be less than 63, and the dimension that can be calculated is restricted by the local RAM condition.
We present a program for solving exactly the general pairing Hamiltonian based on diagonalization. The program generates the seniority-zero shell-model-like basis vectors via the `01' inversion algorithm. The Hamiltonian matrix is constructed in that seniority-zero space. The program evaluates all non-zero elements of the Hamiltonian matrix "on the fly" using the scattering operator and the search algorithm that act on the generated basis. The matrix is diagonalized by using the iterative Lanczos algorithm. The program thus developed, PairDiag, can calculate efficiently the ground-state eigenvalue and eigenvector of any pairing Hamiltonian. The program can be easily implemented to replace the BCS approximation in standard self-consistent mean-field calculations. The code is parallelized using OpenMP. For larger systems with dimension around 108−9^{8-9}8−9, the calculation can be done within a day on standard desktop computers.
We propose a distinct numerical approach to effectively solve the problem of partial diagonalization of the super-large-scale quantum electronic Hamiltonian matrices. The key ingredients of our scheme are the new method for arranging the basis vectors in the computer's RAM and the algorithm allowing not to store a matrix in RAM, but to regenerate it on-the-fly during diagonalization procedure. This scheme was implemented in the program, solving the Anderson impurity model in the framework of dynamical mean-field theory (DMFT). The DMFT equations for electronic Hamiltonian with 18 effective orbitals that corresponds to the matrix with the dimension of 2.4 * 10^9 were solved on the distributed memory computational cluster.
We develop a program package named QS3 based on the (thick-restart) Lanczos method for analyzing spin-1/2 XXZ-type quantum spin models on spatially uniform/non-uniform lattices near fully polarized states, which can be mapped to dilute hardcore Bose systems. All calculations in QS3, including eigenvalue problems, expectation values for one/two-point spin operators, and static/dynamical spin structure factors, are performed in the symmetry-adapted bases specified by the number N↓ of down spins and the wave number k associated with the translational symmetry without using the bit representation for specifying spin configurations. Because of these treatments, QS3 can support large-scale quantum systems containing more than 1000 sites with dilute N↓. We show the benchmark results of QS3 for the low-energy excitation dispersion of the isotropic Heisenberg model on the 10×10×10 cubic lattice, the static and dynamical spin structure factors of the isotropic Heisenberg model on the 10×10 square lattice, and the open-MP parallelization efficiency on the supercomputer (Ohtaka) based on AMD Epyc 7702 installed at the Institute for the Solid State Physics (ISSP). Theoretical backgrounds and the user interface of QS3 are also described. Program Title: QS3 CPC Library link to program files: https://doi.org/10.17632/yc6jytvf8v.1 Developer's repository link: https://github.com/QS-Cube/ED Licensing provisions: MIT Programming language: Fortran90 External routines/libraries: BLAS, LAPACK Nature of problem: Physical properties (such as total energy, magnetic moment, two-point spin correlation function, and dynamical structure factor) Solution method: Application software based on the full diagonalization method and the exact diagonalization method using the Lanczos and thick-restart Lanczos techniques for quantum spin S=1/2 models such as the XXZ model. Restrictions: Spin S=1/2 systems with U(1) symmetry. Unusual features: Massively large quantum spin systems with U(1) symmetry near the saturation field can be solved with or without considering translational symmetry, which is difficult to treat using standard exact diagonalization libraries with the bit representation for specifying spin configurations.
A method is described to construct the ground state of a general pairing Hamiltonian for a finite nuclear systemas a product of collective, real, distinct pairs. These are determined sequentially via an iterative variational procedure that resorts to diagonalizations of the Hamiltonian in restricted model spaces. Different applications of themethod are provided that include comparisonswith exact and projected BCS results. The quantities that are examined are correlation energies, occupation numbers and pair transfer matrix elements. The method is seen to work well leading, under some circumstances, to solutions that are basically exact.
We take the Bose-Hubbard model to illustrate exact diagonalization techniques in a pedagogical way. We follow the road of first generating all the basis vectors, then setting up the Hamiltonian matrix with respect to this basis, and finally using the Lanczos algorithm to solve low lying eigenstates and eigenvalues. Emphasis is placed on how to enumerate all the basis vectors and how to use the hashing trick to set up the Hamiltonian matrix or matrices corresponding to other quantities. Although our route is not necessarily the most efficient one in practice, the techniques and ideas introduced are quite general and may find use in many other problems.
Background: Ab initio many-body methods have been developed over the past ten years to address mid-mass nuclei. In their best current level of implementation, their accuracy is of the order of a few percent error on the ground-state correlation energy. Recently implemented variants of these methods are operating a breakthrough in the description of medium-mass open-shell nuclei at a polynomial computational cost while putting state-of-the-art models of internucleon interactions to the test.Purpose: As progress in the design of internucleon interactions is made, and as questions one wishes to answer are refined in connection with increasingly available experimental data, further efforts must be made to tailor many-body methods that can reach an even higher precision for an even larger number of observable quantum states or nuclei. The objective of the present work is to contribute to such a quest by designing and testing a new many-body scheme.Methods: We formulate a truncated configuration-interaction method that consists of diagonalizing the Hamiltonian in a highly truncated subspace of the total NNN-body Hilbert space. The reduced Hilbert space is generated via the particle-number projected BCS state along with projected seniority-zero two- and four-quasiparticle excitations. Furthermore, the extent by which the underlying BCS state breaks U(1)\text{U}(1)U(1) symmetry is optimized in the presence of the projected two- and four-quasiparticle excitations. This constitutes an extension of the so-called restricted variation after projection method in use within the frame of multireference energy density functional calculations. The quality of the newly designed method is tested against exact solutions of the so-called attractive pairing Hamiltonian problem.Results: By construction, the method reproduces exact results for N=2N=2N=2 and N=4N=4N=4. For N=(8,16,20)N=(8,\phantom{\rule{0.16em}{0ex}}16,\phantom{\rule{0.16em}{0ex}}20)N=(8,16,20), the error in the ground-state correlation energy is less than (0.006%, 0.1%, 0.15%) across the entire range of internucleon coupling defining the pairing Hamiltonian and driving the normal-to-superfluid quantum phase transition. The presently proposed method offers the advantage of automatic access to the low-lying spectroscopy, which it does with high accuracy.Conclusions: The numerical cost of the newly designed variational method is polynomial (N6{N}^{6}N6) in system size. This method achieves unprecedented accuracy for the ground-state correlation energy, effective pairing gap, and one-body entropy as well as for the excitation energy of low-lying states of the attractive pairing Hamiltonian. This constitutes a sufficiently strong motivation to envision its application to realistic nuclear Hamiltonians in view of providing a complementary, accurate, and versatile ab initio description of mid-mass open-shell nuclei in the future.