Transformer-Based Neural Networks Backflow for Strongly Correlated Electronic Structure

• URL: http://arxiv.org/abs/2509.25720v1
• Date: Tue, 30 Sep 2025 03:24:33 GMT
• Title: Transformer-Based Neural Networks Backflow for Strongly Correlated Electronic Structure
• Authors: Huan Ma, Bowen Kan, Honghui Shang, Jinlong Yang,
• Abstract summary: We show that Transformer architectures can be adapted to capture the complex grammar of electronic correlations through neural network backflow.<n>Applying to strongly correlated iron-sulfur clusters validates our approach.<n>Results establish Transformer-based backflow as a powerful variational ansatz for strongly correlated electronic structure.
• Score: 10.809203250926721
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Solving the electronic Schr\"odinger equation for strongly correlated systems remains one of the grand challenges in quantum chemistry. Here we demonstrate that Transformer architectures can be adapted to capture the complex grammar of electronic correlations through neural network backflow. In this approach, electronic configurations are processed as token sequences, where attention layers learn non-local orbital correlations and token-specific neural networks map these contextual representations into backflowed orbitals. Application to strongly correlated iron-sulfur clusters validates our approach: for $\left[\mathrm{Fe}_2 \mathrm{~S}_2\left(\mathrm{SCH}_3\right)_4\right]^{2-}$ ([2Fe-2S]) (30e,20o), the ground-state energy within chemical accuracy of DMRG while predicting magnetic exchange coupling constants closer to experimental values than all compared methods including DMRG, CCSD(T), and recent neural network approaches. For $\left[\mathrm{Fe}_4 \mathrm{S}_4\left(\mathrm{SCH}_3\right)_4\right]^{2-}$ ([4Fe-4S]) (54e,36o), we match DMRG energies and accurately reproduce detailed spin-spin correlation patterns between all Fe centers. The approach scales favorably to large active spaces inaccessible to exact methods, with distributed VMC optimization enabling stable convergence. These results establish Transformer-based backflow as a powerful variational ansatz for strongly correlated electronic structure, achieving superior magnetic property predictions while maintaining chemical accuracy in total energies.

Related papers

• Computing Exchange Coupling constants in Transition metal complexes with Tensor Product Selected Configuration Interaction [0.0]
  Transition metal complexes present significant challenges for electronic structure theory.<n>We compare our recently developed Product Selected configuration Interaction (TPSCI) with Density Matrix Renormalization Group (DMRG) for computing exchange coupling constants in six transition metal systems.<n>Key advantages include natural multistate capability enabling direct J extrapolation with smaller statistical errors, and computational efficiency for challenging systems.
  arXiv  Detail & Related papers  (2025-08-18T15:23:53Z)
• Efficient Prediction of SO(3)-Equivariant Hamiltonian Matrices via SO(2) Local Frames [59.87385171177885]
  We consider the task of predicting Hamiltonian matrices to accelerate electronic structure calculations.<n>Motivated by the inherent relationship between the off-diagonal blocks of the Hamiltonian matrix and the SO(2) local frame, we propose QHNetV2.
  arXiv  Detail & Related papers  (2025-06-11T05:04:29Z)
• Accurate Ab-initio Neural-network Solutions to Large-Scale Electronic Structure Problems [52.19558333652367]
  We present finite-range embeddings (FiRE) for accurate large-scale ab-initio electronic structure calculations.<n>FiRE reduces the complexity of neural-network variational Monte Carlo (NN-VMC) by $sim ntextel$, the number of electrons.<n>We validate our method's accuracy on various challenging systems, including biochemical compounds and organometallic compounds.
  arXiv  Detail & Related papers  (2025-04-08T14:28:54Z)
• Learning Equivariant Non-Local Electron Density Functionals [51.721844709174206]
  We introduce Equivariant Graph Exchange Correlation (EG-XC), a novel non-local XC functional based on equivariant graph neural networks (GNNs)<n>By applying an equivariant GNN on this point cloud, we capture molecular-range interactions in a scalable and accurate manner.<n> EG-XC excels in data efficiency and molecular size extrapolation on QM9, matching force fields trained on 5 times more and larger molecules.
  arXiv  Detail & Related papers  (2024-10-10T14:31:45Z)
• Simple Fermionic backflow states via a systematically improvable tensor decomposition [0.0]
  We present an effective ansatz for the wave function of correlated electrons that brings closer the fields of machine learning parameterizations and tensor rank decompositions. We consider a CANDECOMP/PARAFAC (CP) tensor factorization of a general backflow transformation in second quantization for a simple, compact and systematically improvable Fermionic state.
  arXiv  Detail & Related papers  (2024-07-16T14:37:20Z)
• Machine learning in and out of equilibrium [58.88325379746631]
  Our study uses a Fokker-Planck approach, adapted from statistical physics, to explore these parallels. We focus in particular on the stationary state of the system in the long-time limit, which in conventional SGD is out of equilibrium. We propose a new variation of Langevin dynamics (SGLD) that harnesses without replacement minibatching.
  arXiv  Detail & Related papers  (2023-06-06T09:12:49Z)
• Message-Passing Neural Quantum States for the Homogeneous Electron Gas [41.94295877935867]
  We introduce a message-passing-neural-network-based wave function Ansatz to simulate extended, strongly interacting fermions in continuous space. We demonstrate its accuracy by simulating the ground state of the homogeneous electron gas in three spatial dimensions.
  arXiv  Detail & Related papers  (2023-05-12T04:12:04Z)
• Tensor Factorized Hamiltonian Downfolding To Optimize The Scaling Complexity Of The Electronic Correlations Problem on Classical and Quantum Computers [0.3613661942047476]
  We introduce tensor-factorized Hamiltonian downfolding (TFHD) and its quantum analogue, qubitized downfolding (QD)<n>TFHD collapses every high-rank object to rank-2 networks executed in depth-optimal, block-encoded circuits.<n>We demonstrate super-quadratic speedups of expensive quantum chemistry algorithms on both classical and quantum computers.
  arXiv  Detail & Related papers  (2023-03-13T12:15:54Z)
• Disentangling Interacting Systems with Fermionic Gaussian Circuits: Application to Quantum Impurity Models [0.0]
  We introduce a change of basis with unitary gates obtained from compressing fermionic Gaussian states into quantum circuits corresponding to various tensor networks.<n>These circuits can reduce the ground state entanglement entropy and improve the performance of algorithms such as the density matrix renormalization group.
  arXiv  Detail & Related papers  (2022-12-19T19:11:16Z)
• Simulating Spin-Orbit Coupling With Quasidegenerate N-Electron Valence Perturbation Theory [77.34726150561087]
  We present the first implementation of spin-orbit coupling effects in SO-QDNEVPT2. The accuracy of these methods is tested for the group 14 and 16 hydrides, 3d and 4d transition metal ions, and two actinide dioxides.
  arXiv  Detail & Related papers  (2022-11-11T20:03:37Z)
• Spin Current Density Functional Theory of the Quantum Spin-Hall Phase [59.50307752165016]
  We apply the spin current density functional theory to the quantum spin-Hall phase. We show that the explicit account of spin currents in the electron-electron potential of the SCDFT is key to the appearance of a Dirac cone.
  arXiv  Detail & Related papers  (2022-08-29T20:46:26Z)
• Wave function Ansatz (but Periodic) Networks and the Homogeneous Electron Gas [1.7944372791281356]
  We design a neural network Ansatz for variationally finding the ground-state wave function of the Homogeneous Electron Gas. We study the spin-polarised and paramagnetic phases with 7, 14 and 19 electrons over a broad range of densities. This contribution establishes neural network models as flexible and high precision Ans"atze for periodic electronic systems.
  arXiv  Detail & Related papers  (2022-02-02T14:12:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.