A neural-network-based Python package for performing large-scale atomic CI using pCI and other high-performance atomic codes

Pavlo Bilous; Charles Cheung; Marianna Safronova

Home/Archive/arXiv

arXiv · 2503.01379

A neural-network-based Python package for performing large-scale atomic CI using pCI and other high-performance atomic codes

Pavlo Bilous·Charles Cheung·Marianna Safronova

View original source ↗Read source document ↗Export metadata ↓

Abstract

Modern atomic physics applications in science and technology pose ever higher demands on the precision of computations of properties of atoms and ions. Especially challenging is the modeling of electronic correlations within the configuration interaction (CI) framework, which often requires expansions of the atomic state in huge bases of Slater determinants or configuration state functions. This can easily render the problem intractable even for highly efficient atomic codes running on distributed supercomputer systems. Recently, we have successfully addressed this problem using a neural-network (NN) approach [1]. In this work, we present our Python code for performing NN-supported large-scale atomic CI using pCI [2] and other high-performance atomic codes. [1] P. Bilous, C. Cheung, and M. Safronova, Phys. Rev. A 110 042818 (2024). [2] C. Cheung, M. G. Kozlov, S. G. Porsev, M. S. Safronova, I. I. Tupitsyn, A. I. Bondarev, Comput. Phys. Commun. 308 109463 (2025).

Explore related subjects

physics.atom-ph

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Pavlo Bilous, Charles Cheung, Marianna Safronova. 2025-03-03. A neural-network-based Python package for performing large-scale atomic CI using pCI and other high-performance atomic codes. https://arxiv.org/abs/2503.01379

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

State-Dependent Diffusion and Spectra of Strongly Driven Thermal Atoms

We propose a state-dependent diffusion model for strongly driven thermal-atom spectra. Starting from the trajectory-dependent internal-state evolution of individual atoms, we derive a closed spatial equation for the local density-matrix field using a velocity-moment expansion. Measurements of an $^{85}$Rb atomic-filter transmission spectrum agree well with the model up to a maximum Gaussian peak intensity of $1.27\times10^{3}$ W/cm$^2$, approaching six orders of magnitude above the $^{85}$Rb D2-line saturation intensity. Counterintuitively, the model reveals an anomalous optical-pumping pathway in which intense light transfers atoms from nominally dark states into bright states. Hyperfine Paschen--Back splitting selectively enhances this anomalous pathway while suppressing conventional optical pumping, allowing the filter to maintain approximately 97$\%$ transmission at the highest intensity studied. This work provides a framework for controlling strongly driven atomic ensembles and designing saturation-resistant atomic optical devices.

physics.atom-ph↗

arXiv2026

Interactions and excitation blockade of Rydberg macrodimers

We present a theoretical study of interactions between Rydberg macrodimers, molecular bound states of Rydberg atoms, and the largest known neutral homonuclear diatomic molecules. Accounting for the large number of molecular interaction channels from four interacting Rydberg atoms, we show that macrodimer-macrodimer interactions are anisotropic and similar in strength and range to atomic-state Rydberg interactions. Our analysis reveals the presence of two-atom, three-atom, and four-atom contributions in the bimolecular system, and elucidates how these interactions are rooted in the underlying atomic Rydberg-state couplings. Our calculations of the excitation blockade between macrodimers are in reasonable agreement with experimentally observed signals, and pave the way for the exploration of optically controlled collective quantum dynamics in molecular systems.

physics.atom-ph↗

arXiv2026

Non-orthogonal extension of Graspg - dynamic electron correlation for large and compact active spaces

Accurate relativistic multiconfiguration calculations of correlation-sensitive atomic properties are often limited by the rapid growth of configuration state function expansions when a single common orthonormal orbital basis is used. In this work, a partitioned correlation function interaction (PCFI) method is developed for relativistic atomic structure calculations. The correlation space is separated into physically motivated components, which are optimized independently with correlation-specific orbital sets. The interactions between configuration spaces constructed from mutually non-orthogonal orbital sets are evaluated using biorthonormal transformations, allowing different correlation effects to be combined in a compact final interaction calculation. Full details of the method are provided, emphasizing its connection to configuration state function generators (CSFGs), which significantly reduce the time required to construct the Hamiltonian matrix in conventional RCI calculations. Applications to the neutral Li, Be, and Al atoms are presented for energy levels, mass shifts and hyperfine structure constants. Compared with conventional relativistic configuration interaction (RCI) calculations that rely on a single orbital basis, PCFI produces more compact and predictable convergence patterns for both total and transition energies. It also offers greater stability for correlation-sensitive properties such as specific mass shifts and hyperfine constants. By using property-oriented partitions, PCFI captures core-polarization effects more effectively, thereby reducing the oscillatory behavior often observed in standard RCI approaches. Overall, the results demonstrate that PCFI provides a promising and computationally efficient framework for accurate relativistic multiconfiguration calculations of correlation-dependent atomic properties.

physics.atom-ph↗