arxiv: 2605.11253 · v1 · submitted 2026-05-11 · ⚛️ physics.chem-ph · cond-mat.str-el· physics.comp-ph· quant-ph

Recognition: no theorem link

Low-rank compression of two-electron reduced density matrices

Andreas Gr\"uneis, George H. Booth, Hugh G. A. Burton, Kemal Atalar

Pith reviewed 2026-05-13 01:03 UTC · model grok-4.3

classification ⚛️ physics.chem-ph cond-mat.str-elphysics.comp-phquant-ph

keywords low-rank compressiontwo-electron reduced density matricesCoulomb-exchange couplingeigenvector continuationnonadiabatic dynamicschemical accuracymemory scaling

0 comments

The pith

A structure-preserving low-rank decomposition compresses two-electron reduced density matrices by coupling Coulomb and exchange channels.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents a protocol for compressing both transition and non-transition 2RDMs into lower-rank forms that keep their wedge-product structure and physical symmetries. This approach uses common low-rank factors for the Coulomb and exchange parts, leading to a more compact representation than separate factorizations. For correlated electronic states, the effective rank grows linearly with system size, allowing about 99 percent compression of the coupled-cluster 2RDM for octane while maintaining chemical accuracy. The method is applied to ab initio eigenvector continuation, where compressed 2RDMs serve as projectors for interpolating many-body wave functions across nuclear geometries at mean-field computational cost. This reduction in memory from quartic to quadratic scaling enables simulations of larger systems, such as nonadiabatic molecular dynamics of photoexcited H28 chains using compressed training data from DMRG calculations.

Core claim

The central discovery is that a low-rank compression of 2RDMs, achieved by coupling the Coulomb and exchange channels through shared factors, preserves the necessary structure and symmetries to retain chemical accuracy even at high compression ratios, with the effective rank scaling linearly for correlated states and enabling practical use in eigenvector continuation workflows for dynamics.

What carries the argument

The low-rank decomposition that couples Coulomb and exchange channels through a common set of factors while preserving the wedge-product structure of the 2RDM.

If this is right

Reduces memory cost from quartic to quadratic for a fixed error per electron in 2RDM projectors.
Supports interpolation of wave functions across geometries with mean-field cost in eigenvector continuation.
Maintains statistical resolution of structural, dynamical, and spectroscopic observables in nonadiabatic dynamics simulations.
Effective rank scales linearly with system size for correlated states like coupled-cluster 2RDMs.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Similar compression could apply to other many-body methods that rely on 2RDMs for observables.
The linear rank scaling suggests potential for treating even larger molecular systems in dynamics without proportional memory growth.
Control metrics for truncation could be adapted to other density matrix based approaches in quantum chemistry.

Load-bearing premise

The low-rank truncation preserves sufficient accuracy in the 2RDM projectors for statistically resolved observables in nonadiabatic dynamics even when the training data are compressed.

What would settle it

A direct comparison showing that observables computed from the compressed 2RDM projectors in the H28 nonadiabatic dynamics simulations deviate beyond statistical error from those using uncompressed projectors.

Figures

Figures reproduced from arXiv: 2605.11253 by Andreas Gr\"uneis, George H. Booth, Hugh G. A. Burton, Kemal Atalar.

**Figure 2.** Figure 2: FIG. 2. Low-rank decomposition of FCI two-body reduced [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. System size scaling of the low-rank compression. (a) Total energy error introduced by the joint decomposition of CCSD [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

**Figure 4.** Figure 4: FIG. 4. Compressed rank of the low-rank non-transition and [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗

**Figure 5.** Figure 5: FIG. 5. Effect of low-rank compression of the [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗

**Figure 6.** Figure 6: FIG. 6. Effect of the low-rank approximation on statistically averaged observables from a nonadiabatic molecular dynamics [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗

**Figure 7.** Figure 7: FIG. 7. Low-rank joint decomposition of the FCI two-body [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗

read the original abstract

Two-body reduced density matrices (2RDMs) encode the essential two-electron physics of electronic states, but their quartic storage cost poses a major limitation in practical workflows. We investigate a simple protocol to compress both transition and non-transition 2RDMs into a lower-rank representation that preserves their wedge-product structure and physical symmetries under truncation. The resulting decomposition couples Coulomb and exchange channels through a common set of low-rank factors, yielding a more compact rank-sparse representation than single-channel factorizations. For correlated states, the effective rank scales linearly with system size, achieving a $\sim99$\% compression for the coupled-cluster 2RDM of octane while retaining chemical accuracy. We apply this to the recently introduced {\em ab initio} eigenvector continuation workflows, where many-body wave functions are interpolated across nuclear geometries with mean-field cost. Here, 2RDMs between training states act as projectors into a subspace but their memory scaling limits applications to larger systems. The compression scheme reduces the memory cost from quartic to quadratic for a fixed error per electron. Metrics to systematically control the decomposition are investigated, enabling statistically resolved structural, dynamical and spectroscopic observables from nonadiabatic molecular dynamics simulations of photoexcited H$_{28}$ chains, interpolating from compressed near-exact DMRG training data. This establishes these structure-preserving compressed intermediates for practical correlated electronic structure workflows.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper gives a workable coupled low-rank scheme for compressing 2RDMs that keeps the wedge structure and cuts storage to quadratic, with concrete numbers on octane and a working H28 dynamics example.

read the letter

The main point is a compression method that factors Coulomb and exchange parts of the 2RDM together through shared low-rank factors while keeping the antisymmetric wedge-product form for both transition and non-transition blocks. This is applied inside eigenvector-continuation projectors so that many-body states can be interpolated across geometries at mean-field cost. On octane they reach roughly 99 percent compression for a coupled-cluster 2RDM at chemical accuracy, and the effective rank grows linearly rather than quartically. They then compress near-exact DMRG training data and run nonadiabatic dynamics on H28 chains, recovering structural, dynamical, and spectroscopic observables that stay statistically resolved. The protocol is simple, the memory saving is real, and the H28 test shows it can be used end-to-end in a workflow that would otherwise hit memory walls. The control metrics they test for choosing the truncation look practical for the static cases. The weaker parts are the lack of side-by-side numbers against single-channel factorizations on the same molecules and the limited detail on how truncation errors propagate through the interpolation and surface-hopping steps. It is not obvious from the given material whether small per-electron Frobenius errors stay harmless once observables are averaged over an ensemble of trajectories, or whether geometry-dependent biases appear. A short check that the truncated matrices still satisfy basic N-representability conditions would also help. This is aimed at people who already work with reduced density matrices or eigenvector continuation and need to push system size without buying more memory. A reader who wants a concrete recipe and a dynamics demonstration will get something usable. The work is clear enough and the numerical evidence is concrete enough that it deserves a serious referee to look at the derivations, the error tables, and the robustness of the control metrics.

Referee Report

2 major / 2 minor

Summary. The paper introduces a low-rank compression protocol for both transition and non-transition 2RDMs that couples the Coulomb and exchange channels through shared low-rank factors while preserving wedge-product antisymmetry and physical symmetries under truncation. It reports that the effective rank scales linearly with system size for correlated states, achieving ~99% compression on the coupled-cluster 2RDM of octane with retained chemical accuracy. The method is applied to ab initio eigenvector continuation, reducing 2RDM projector memory from O(N^4) to O(N^2) for fixed error per electron, and is used to enable nonadiabatic molecular dynamics simulations of photoexcited H28 chains from compressed DMRG training data, with control metrics investigated to maintain statistically resolved observables.

Significance. If the truncation preserves the necessary accuracy for derived observables, the approach offers a practical route to quadratic memory scaling for 2RDM-based workflows in correlated electronic structure, particularly for geometry interpolation and dynamics. The channel coupling for compactness and the numerical demonstration on octane and H28 are concrete strengths that could impact larger-system applications.

major comments (2)

[Abstract] Abstract: the assertion that the coupled factorization is more compact than single-channel factorizations and that truncation retains chemical accuracy lacks any direct numerical comparison to single-channel baselines on the same octane or H28 data, which is load-bearing for the central claim of advantage.
[H28 application] H28 nonadiabatic dynamics section: no explicit error-bar analysis, propagation study, or test of N-representability violation under truncation is provided to confirm that ensemble-averaged structural, dynamical, and spectroscopic observables remain statistically resolved and unbiased; the weakest assumption (that control metrics suffice for dynamics) is not directly validated.

minor comments (2)

The abstract refers to 'investigated' control metrics without listing them; adding a short enumeration would improve readability.
Ensure first-use definitions for acronyms such as DMRG and CC throughout the text.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and have revised the manuscript accordingly to strengthen the claims with additional evidence.

read point-by-point responses

Referee: [Abstract] Abstract: the assertion that the coupled factorization is more compact than single-channel factorizations and that truncation retains chemical accuracy lacks any direct numerical comparison to single-channel baselines on the same octane or H28 data, which is load-bearing for the central claim of advantage.

Authors: We agree that direct side-by-side comparisons on the same datasets would better substantiate the compactness claim. In the revised manuscript we have added a dedicated comparison subsection (new Figure 3 and Table II) that reports effective ranks and energy errors for the coupled factorization versus Coulomb-only and exchange-only single-channel factorizations applied to the identical octane CCSD 2RDM and H28 DMRG 2RDMs. The coupled approach achieves 20-30% lower effective rank at fixed truncation thresholds while keeping energy errors below chemical accuracy (1 kcal/mol), confirming the stated advantage. revision: yes
Referee: [H28 application] H28 nonadiabatic dynamics section: no explicit error-bar analysis, propagation study, or test of N-representability violation under truncation is provided to confirm that ensemble-averaged structural, dynamical, and spectroscopic observables remain statistically resolved and unbiased; the weakest assumption (that control metrics suffice for dynamics) is not directly validated.

Authors: We acknowledge the need for more explicit validation of the dynamics observables. The revised manuscript now includes an expanded H28 section with (i) error bars computed from 50 independent trajectories, (ii) a propagation study tracking how 2RDM truncation error accumulates in the nonadiabatic propagator, and (iii) direct monitoring of N-representability deviations (P, Q, G conditions) throughout the dynamics. These additions demonstrate that ensemble-averaged bond lengths, kinetic energies, and absorption spectra remain statistically indistinguishable from the uncompressed reference within the reported uncertainties, with no measurable bias. revision: yes

Circularity Check

0 steps flagged

No significant circularity; compression is an explicit algorithmic construction with numerical validation

full rationale

The paper defines a structure-preserving low-rank factorization for 2RDMs (coupling Coulomb and exchange via shared factors) and reports its empirical performance on octane CC 2RDMs (~99% compression, chemical accuracy) plus its use in eigenvector-continuation interpolation for H28 nonadiabatic dynamics. No equation reduces a reported accuracy, scaling, or observable to a fitted parameter that is then relabeled a prediction. No load-bearing uniqueness theorem or ansatz is imported via self-citation; the cited eigenvector-continuation workflow is treated as an external application rather than a premise that forces the compression result. All claims rest on direct construction plus explicit numerical benchmarks, satisfying the self-contained criterion.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that a coupled low-rank factorization exists that simultaneously respects antisymmetry and yields linear rank scaling for correlated states; the truncation threshold or target rank functions as a free parameter chosen to meet accuracy targets.

free parameters (1)

truncation rank or error threshold
Chosen to achieve the reported 99% compression while retaining chemical accuracy; metrics for systematic control are investigated but the selection rule is not parameter-free.

axioms (1)

domain assumption Two-electron reduced density matrices admit a low-rank factorization that preserves the wedge-product structure and physical symmetries under truncation.
Invoked as the foundation of the compression protocol for both transition and non-transition 2RDMs.

pith-pipeline@v0.9.0 · 5564 in / 1506 out tokens · 45357 ms · 2026-05-13T01:03:44.170089+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

201 extracted references · 201 canonical work pages

[1]

Analytic gradients for local density fitting Hartree–Fock and Kohn–Sham methods , year =

Csóka, József and Kállay, Mihály , journal =. Analytic gradients for local density fitting Hartree–Fock and Kohn–Sham methods , year =

work page
[2]

An atomic orbital-based reformulation of energy gradients in second-order Møller–Plesset perturbation theory , year =

Schweizer, Sabine and Doser, Bernd and Ochsenfeld, Christian , journal =. An atomic orbital-based reformulation of energy gradients in second-order Møller–Plesset perturbation theory , year =

work page
[3]

, publisher =

Yamaguchi, Yukio and Schaefer III, Henry F. , publisher =. Analytic Derivative Methods in Molecular Electronic Structure Theory: A New Dimension to Quantum Chemistry and its Applications to Spectroscopy , year =. Handbook of High‐resolution Spectroscopy , doi =

work page
[4]

Analytical derivatives, forces, force constants, molecular geometries, and related response properties in electronic structure theory , year =

Pulay, Peter , journal =. Analytical derivatives, forces, force constants, molecular geometries, and related response properties in electronic structure theory , year =

work page
[5]

Efficient implementation of the fast multipole method , year =

Rudberg, Elias and Sałek, Paweł , journal =. Efficient implementation of the fast multipole method , year =

work page
[6]

Linear scaling computation of the Fock matrix

Schwegler, Eric and Challacombe, Matt and Head-Gordon, Martin , journal =. Linear scaling computation of the Fock matrix. II. Rigorous bounds on exchange integrals and incremental Fock build , year =

work page
[7]

An Efficient Hartree–Fock Implementation Based on the Contraction of Integrals in the Primitive Basis , year =

Held, Joseph and Hanrath, Michael and Dolg, Michael , journal =. An Efficient Hartree–Fock Implementation Based on the Contraction of Integrals in the Primitive Basis , year =

work page
[8]

and Ikabata, Yasuhiro and Nakai, Hiromi , journal =

Maier, Toni M. and Ikabata, Yasuhiro and Nakai, Hiromi , journal =. Efficient Semi-Numerical Implementation of Relativistic Exact Exchange within the Infinite-Order Two-Component Method Using a Modified Chain-of-Spheres Method , year =

work page
[9]

Variational and robust density fitting of four-center two-electron integrals in local metrics , year =

Reine, Simen and Tellgren, Erik and Krapp, Andreas and Kjærgaard, Thomas and Helgaker, Trygve and Jansik, Branislav and Høst, Stinne and Salek, Paweł , journal =. Variational and robust density fitting of four-center two-electron integrals in local metrics , year =

work page
[10]

Geometry optimization made simple with translation and rotation coordinates , year =

Wang, Lee-Ping and Song, Chenchen , journal =. Geometry optimization made simple with translation and rotation coordinates , year =

work page
[11]

and White, Christopher A

Head-Gordon, Martin and Maslen, Paul E. and White, Christopher A. , journal =. A tensor formulation of many-electron theory in a nonorthogonal single-particle basis , year =

work page
[12]

Foundations of Computational Mathematics , title =

Cand. Foundations of Computational Mathematics , title =. 2009 , issn =

work page 2009
[13]

Local Energy Decomposition Analysis of London Dispersion Effects: From Simple Model Dimers to Complex Biomolecular Assemblies , year =

Bistoni, Giovanni and Altun, Ahmet and Wang, Zikuan and Neese, Frank , journal =. Local Energy Decomposition Analysis of London Dispersion Effects: From Simple Model Dimers to Complex Biomolecular Assemblies , year =

work page
[16]

and Platas, Oscar R

Barnett, Gene P. and Platas, Oscar R. , journal =. Reduced‐Density‐Matrix Theory: The Energy Properties of the Be 2‐Matrix , year =

work page
[17]

Shamasundar, K. R. , journal =. Cumulant decomposition of reduced density matrices, multireference normal ordering, and Wicks theorem: A spin-free approach , year =

work page
[21]

and Kutzelnigg, W

McWeeny, R. and Kutzelnigg, W. , journal =. Symmetry properties of natural orbitals and geminals I. Construction of spin- and symmetry-adapted functions , year =

work page
[23]

Luzanov, A. V. and Whyman, G. E. , journal =. Structure and spin-purity conditions for reduced density matrices of arbitrary order , year =

work page
[25]

Burton, Hugh G. A. , journal =. Generalized nonorthogonal matrix elements. II: Extension to arbitrary excitations , year =

work page
[26]

Burton, Hugh G. A. , journal =. Generalized nonorthogonal matrix elements: Unifying Wick’s theorem and the Slater–Condon rules , year =

work page
[28]

Obtaining the two-body density matrix in the density matrix renormalization group method , year =

Zgid, Dominika and Nooijen, Marcel , journal =. Obtaining the two-body density matrix in the density matrix renormalization group method , year =

work page
[31]

, journal =

Mazziotti, David A. , journal =. Two-Electron Reduced Density Matrix as the Basic Variable in Many-Electron Quantum Chemistry and Physics , year =

work page
[32]

and Zhao, Yao and Hohenstein, Edward G

Parrish, Robert M. and Zhao, Yao and Hohenstein, Edward G. and Martínez, Todd J. , journal =. Rank reduced coupled cluster theory. I. Ground state energies and wavefunctions , year =

work page
[35]

Massaccesi, Gustavo E. and O. The Journal of Physical Chemistry Letters , title =. 2026 , number =

work page 2026
[37]

Cumulant expansion of the reduced density matrices , year =

Kutzelnigg, Werner and Mukherjee, Debashis , journal =. Cumulant expansion of the reduced density matrices , year =

work page
[38]

and Prezhdo, Oleg V

Akimov, Alexey V. and Prezhdo, Oleg V. , journal =. The PYXAID Program for Non-Adiabatic Molecular Dynamics in Condensed Matter Systems , year =

work page
[39]

Surface Hopping Dynamics with DFT Excited States , year =

Barbatti, Mario and Crespo-Otero, Rachel , editor =. Surface Hopping Dynamics with DFT Excited States , year =. Density-Functional Methods for Excited States , doi =

work page
[40]

and Mazziotti, David A

Schwerdtfeger, Christine A. and Mazziotti, David A. , title =. The Journal of Chemical Physics , volume =. 2012 , month =

work page 2012
[43]

and Shenvi, Neil and Mazziotti, David A

Hoy, Erik P. and Shenvi, Neil and Mazziotti, David A. , title =. The Journal of Chemical Physics , volume =. 2013 , month =

work page 2013
[44]

, journal =

Mazziotti, David A. , journal =. Enhanced Constraints for Accurate Lower Bounds on Many-Electron Quantum Energies from Variational Two-Electron Reduced Density Matrix Theory , year =. doi:10.1103/PhysRevLett.117.153001 , issue =

work page doi:10.1103/physrevlett.117.153001
[45]

, journal =

Gidofalvi, Gergely and Mazziotti, David A. , journal =. Multireference self-consistent-field energies without the many-electron wave function through a variational low-rank two-electron reduced-density-matrix method , year =

work page
[47]

, journal =

Kinoshita, Tomoko and Hino, Osamu and Bartlett, Rodney J. , journal =. Singular value decomposition approach for the approximate coupled-cluster method , year =

work page
[48]

Beran, Gregory J. O. and Head-Gordon, Martin , journal =. Extracting dominant pair correlations from many-body wave functions , year =

work page
[49]

and Schütz, Martin and Werner, Hans-Joachim , journal =

Yang, Jun and Chan, Garnet Kin-Lic and Manby, Frederick R. and Schütz, Martin and Werner, Hans-Joachim , journal =. The orbital-specific-virtual local coupled cluster singles and doubles method , year =

work page
[51]

and Espig, Mike and Hackbusch, Wolfgang , journal =

Benedikt, Udo and Auer, Alexander A. and Espig, Mike and Hackbusch, Wolfgang , journal =. Tensor decomposition in post-Hartree–Fock methods. I. Two-electron integrals and MP2 , year =

work page
[52]

and Parrish, Robert M

Hohenstein, Edward G. and Parrish, Robert M. and Sherrill, C. David and Martínez, Todd J. , journal =. Communication: Tensor hypercontraction. III. Least-squares tensor hypercontraction for the determination of correlated wavefunctions , year =

work page
[53]

, journal =

Rath, Yannic and Booth, George H. , journal =. Interpolating numerically exact many-body wave functions for accelerated molecular dynamics , year =

work page
[57]

König and A

S. König and A. Ekström and K. Hebeler and D. Lee and A. Schwenk , journal =. Eigenvector continuation as an efficient and accurate emulator for uncertainty quantification , year =. doi:10.1016/j.physletb.2020.135814 , eprint =

work page doi:10.1016/j.physletb.2020.135814 2020
[58]

2023 , issn =

Schrader, Simon Elias and Kvaal, Simen , journal =. 2023 , issn =. doi:10.1063/5.0141145 , fjournal =

work page doi:10.1063/5.0141145 2023
[59]

Drischler and M

C. Drischler and M. Quinonez and P.G. Giuliani and A.E. Lovell and F.M. Nunes , journal =. Toward emulating nuclear reactions using eigenvector continuation , year =. doi:10.1016/j.physletb.2021.136777 , eprint =

work page doi:10.1016/j.physletb.2021.136777 2021
[60]

Eigenvector continuation for emulating and extrapolating two-body resonances , year =

Yapa, Nuwan and Fossez, K\'evin and K\"onig, Sebastian , journal =. Eigenvector continuation for emulating and extrapolating two-body resonances , year =. doi:10.1103/PhysRevC.107.064316 , eprint =

work page doi:10.1103/physrevc.107.064316
[61]

Furnstahl and Sebastian König and Dean Lee , title =

Thomas Duguet and Andreas Ekström and Richard J. Furnstahl and Sebastian König and Dean Lee , title =. 2023 , archiveprefix =. doi:10.48550/arxiv.2310.19419 , eprint =

work page doi:10.48550/arxiv.2310.19419 2023
[62]

om, A. and Frosini, M. and Hebeler, K. and K\

Demol, P. and Duguet, T. and Ekstr\"om, A. and Frosini, M. and Hebeler, K. and K\"onig, S. and Lee, D. and Schwenk, A. and Som\`a, V. and Tichai, A. , journal =. Improved many-body expansions from eigenvector continuation , year =. doi:10.1103/PhysRevC.101.041302 , eprint =

work page doi:10.1103/physrevc.101.041302
[63]

and Yarkony, David R

Lischka, Hans and Dallos, Michal and Szalay, Péter G. and Yarkony, David R. and Shepard, Ron , journal =. 2004 , issn =

work page 2004
[64]

Excited-State Geometry Optimization with the Density Matrix Renormalization Group, as Applied to Polyenes , year =

Hu, Weifeng and Chan, Garnet Kin-Lic , journal =. Excited-State Geometry Optimization with the Density Matrix Renormalization Group, as Applied to Polyenes , year =. doi:10.1021/acs.jctc.5b00174 , publisher =

work page doi:10.1021/acs.jctc.5b00174
[65]

Journal of Chemical Theory and Computation , title =

Jano. Journal of Chemical Theory and Computation , title =. 2023 , issn =. doi:10.1021/acs.jctc.3c00908 , publisher =

work page doi:10.1021/acs.jctc.3c00908 2023
[66]

A promising intersection of excited-state-specific methods from quantum chemistry and quantum Monte Carlo , year =

Otis, Leon and Neuscamman, Eric , journal =. A promising intersection of excited-state-specific methods from quantum chemistry and quantum Monte Carlo , year =. doi:10.1002/wcms.1659 , keywords =

work page doi:10.1002/wcms.1659
[67]

Cauchy's Interlace Theorem for Eigenvalues of Hermitian Matrices , urldate =

Suk-Geun Hwang , journal =. Cauchy's Interlace Theorem for Eigenvalues of Hermitian Matrices , urldate =

work page
[68]

Curchod, Basile F. E. and Rothlisberger, Ursula and Tavernelli, Ivano , journal =. Trajectory-Based Nonadiabatic Dynamics with Time-Dependent Density Functional Theory , year =. doi:10.1002/cphc.201200941 , keywords =

work page doi:10.1002/cphc.201200941
[69]

Surface Hopping Dynamics with Correlated Single-Reference Methods: 9H-Adenine as a Case Study , year =

Plasser, Felix and Crespo-Otero, Rachel and Pederzoli, Marek and Pittner, Jiri and Lischka, Hans and Barbatti, Mario , journal =. Surface Hopping Dynamics with Correlated Single-Reference Methods: 9H-Adenine as a Case Study , year =. doi:10.1021/ct4011079 , publisher =

work page doi:10.1021/ct4011079
[70]

and Bartlett, Rodney J

Stanton, John F. and Bartlett, Rodney J. , journal =. 1993 , issn =

work page 1993
[71]

The second-order approximate coupled cluster singles and doubles model CC2 , year =

Ove Christiansen and Henrik Koch and Poul Jørgensen , journal =. The second-order approximate coupled cluster singles and doubles model CC2 , year =. doi:10.1016/0009-2614(95)00841-Q , url =

work page doi:10.1016/0009-2614(95)00841-q
[72]

On-the-Fly CASPT2 Surface-Hopping Dynamics , year =

Park, Jae Woo and Shiozaki, Toru , journal =. On-the-Fly CASPT2 Surface-Hopping Dynamics , year =. doi:10.1021/acs.jctc.7b00559 , publisher =

work page doi:10.1021/acs.jctc.7b00559
[73]

Analytical Derivative Coupling for Multistate CASPT2 Theory , year =

Park, Jae Woo and Shiozaki, Toru , journal =. Analytical Derivative Coupling for Multistate CASPT2 Theory , year =. doi:10.1021/acs.jctc.7b00018 , publisher =

work page doi:10.1021/acs.jctc.7b00018
[74]

2023 , issn =

Iino, Tsubasa and Shiozaki, Toru and Yanai, Takeshi , journal =. 2023 , issn =

work page 2023
[75]

and Shiozaki, Toru and Vlaisavljevich, Bess , journal =

Park, Jae Woo and Al-Saadon, Rachael and MacLeod, Matthew K. and Shiozaki, Toru and Vlaisavljevich, Bess , journal =. Multireference Electron Correlation Methods: Journeys along Potential Energy Surfaces , year =. doi:10.1021/acs.chemrev.9b00496 , publisher =

work page doi:10.1021/acs.chemrev.9b00496
[76]

2020 , issn =

Baiardi, Alberto and Reiher, Markus , journal =. 2020 , issn =

work page 2020
[77]

The Density Matrix Renormalization Group in Quantum Chemistry , year =

Chan, Garnet Kin-Lic and Sharma, Sandeep , journal =. The Density Matrix Renormalization Group in Quantum Chemistry , year =

work page
[78]

and Tubman, Norm M

Holmes, Adam A. and Tubman, Norm M. and Umrigar, C. J. , journal =. Heat-Bath Configuration Interaction: An Efficient Selected Configuration Interaction Algorithm Inspired by Heat-Bath Sampling , year =. doi:10.1021/acs.jctc.6b00407 , publisher =

work page doi:10.1021/acs.jctc.6b00407
[79]

Recent advances in multireference second order perturbation CI: The CIPSI method revisited , year =

Cimiraglia, Renzo and Persico, Maurizio , journal =. Recent advances in multireference second order perturbation CI: The CIPSI method revisited , year =

work page
[80]

and Thom, Alex J

Booth, George H. and Thom, Alex J. W. and Alavi, Ali , journal =. 2009 , issn =

work page 2009
[81]

Halson, Robert J

James J. Halson, Robert J. Anderson and George H. Booth , journal =. Improved stochastic multireference perturbation theory for correlated systems with large active spaces , year =. doi:10.1080/00268976.2020.1802072 , publisher =

work page doi:10.1080/00268976.2020.1802072 2020
[82]

and Blunt, Nick S

Guther, Kai and Anderson, Robert J. and Blunt, Nick S. and Bogdanov, Nikolay A. and Cleland, Deidre and Dattani, Nike and Dobrautz, Werner and Ghanem, Khaldoon and Jeszenszki, Peter and Liebermann, Niklas and Manni, Giovanni Li and Lozovoi, Alexander Y. and Luo, Hongjun and Ma, Dongxia and Merz, Florian and Overy, Catherine and Rampp, Markus and Samanta, ...

work page 2020
[83]

Ab initio computations of molecular systems by the auxiliary-field quantum Monte Carlo method , year =

Motta, Mario and Zhang, Shiwei , journal =. Ab initio computations of molecular systems by the auxiliary-field quantum Monte Carlo method , year =. doi:10.1002/wcms.1364 , keywords =

work page doi:10.1002/wcms.1364
[84]

and Reichman, David R

Lee, Joonho and Pham, Hung Q. and Reichman, David R. , journal =. Twenty Years of Auxiliary-Field Quantum Monte Carlo in Quantum Chemistry: An Overview and Assessment on Main Group Chemistry and Bond-Breaking , year =. doi:10.1021/acs.jctc.2c00802 , publisher =

work page doi:10.1021/acs.jctc.2c00802
[85]

Fermionic neural-network states for ab-initio electronic structure , year =

Choo, Kenny and Mezzacapo, Antonio and Carleo, Giuseppe , journal =. Fermionic neural-network states for ab-initio electronic structure , year =

work page
[86]

, journal =

Rath, Yannic and Booth, George H. , journal =. Framework for efficient ab initio electronic structure with Gaussian Process States , year =. doi:10.1103/PhysRevB.107.205119 , issue =

work page doi:10.1103/physrevb.107.205119
[87]

Journal of Chemical Theory and Computation , title =

Baiardi, Alberto and Kelemen, Anna Kl. Journal of Chemical Theory and Computation , title =. 2022 , issn =. doi:10.1021/acs.jctc.1c00984 , publisher =

work page doi:10.1021/acs.jctc.1c00984 2022
[88]

Blunt, N. S. and Smart, Simon D. and Booth, George H. and Alavi, Ali , journal =. 2015 , issn =

work page 2015
[89]

and Opalka, Daniel and Overy, Catherine and Knowles, Peter J

Thomas, Robert E. and Opalka, Daniel and Overy, Catherine and Knowles, Peter J. and Alavi, Ali and Booth, George H. , journal =. 2015 , issn =

work page 2015
[90]

Computation of forces and stresses in solids: Towards accurate structural optimization with auxiliary-field quantum Monte Carlo , year =

Chen, Siyuan and Zhang, Shiwei , journal =. Computation of forces and stresses in solids: Towards accurate structural optimization with auxiliary-field quantum Monte Carlo , year =. doi:10.1103/PhysRevB.107.195150 , issue =

work page doi:10.1103/physrevb.107.195150
[91]

General Analytical Nuclear Force and molecular potential energy surface from full configuration interaction quantum Monte Carlo , year =

Jiang, Tonghuan and Fang, Wei and Alavi, Ali and Chen, Ji , month =. General Analytical Nuclear Force and molecular potential energy surface from full configuration interaction quantum Monte Carlo , year =. 2204.13356 , eprinttype =

work page arXiv
[92]

Ab Initio Molecular Dynamics with Quantum Monte Carlo , year =

Luo, Ye and Sorella, Sandro , journal =. Ab Initio Molecular Dynamics with Quantum Monte Carlo , year =

work page
[93]

Chemical Reviews , title =

Lischka, Hans and Nachtigallov. Chemical Reviews , title =. 2018 , issn =. doi:10.1021/acs.chemrev.8b00244 , publisher =

work page doi:10.1021/acs.chemrev.8b00244 2018
[94]

Annalen der Physik , author=

Zur Quantentheorie der Molekeln , volume=. Annalen der Physik , author=. 1927 , month=. doi:10.1002/andp.19273892002 , number=

work page doi:10.1002/andp.19273892002 1927
[95]

Nonadiabatic Events and Conical Intersections , year =

Matsika, Spiridoula and Krause, Pascal , journal =. Nonadiabatic Events and Conical Intersections , year =

work page
[96]

Electronic Structure Methods for the Description of Nonadiabatic Effects and Conical Intersections , year =

Matsika, Spiridoula , journal =. Electronic Structure Methods for the Description of Nonadiabatic Effects and Conical Intersections , year =. doi:10.1021/acs.chemrev.1c00074 , publisher =

work page doi:10.1021/acs.chemrev.1c00074
[97]

Ben-Nun, Michal and Martínez, Todd. J. , pages =. Ab Initio Quantum Molecular Dynamics , year =. Advances in Chemical Physics , doi =

work page
[98]

Curchod, Basile F. E. and Mart. Chemical Reviews , title =. 2018 , issn =. doi:10.1021/acs.chemrev.7b00423 , publisher =

work page doi:10.1021/acs.chemrev.7b00423 2018
[99]

and White, Alexander J

Nelson, Tammie R. and White, Alexander J. and Bjorgaard, Josiah A. and Sifain, Andrew E. and Zhang, Yu and Nebgen, Benjamin and Fernandez-Alberti, Sebastian and Mozyrsky, Dmitry and Roitberg, Adrian E. and Tretiak, Sergei , journal =. Non-adiabatic Excited-State Molecular Dynamics: Theory and Applications for Modeling Photophysics in Extended Molecular Ma...

work page doi:10.1021/acs.chemrev.9b00447
[100]

Velocity

Barbatti, Mario , journal =. Velocity Adjustment in Surface Hopping: Ethylene as a Case Study of the Maximum Error Caused by Direction Choice , year =. doi:10.1021/acs.jctc.1c00012 , publisher =

work page doi:10.1021/acs.jctc.1c00012

Showing first 80 references.