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

Generalized Neural Wave Functions

Nicholas Gao, Stephan Günnemann, Technische Universität München

Published Jan 1, 2025• 0 citations
OpenAlex

Related papers

Excited Pfaffians: A Masterclass on Generalized Neural Wave Functions Across Structure and State

Ijaz Durrani

•Jan 1, 2026•0 citations

Convolutional Neural Network Wave Functions: learning quantum many-body physics

Douglas Hendry, Adrian E. Feiguin

Bulletin of the American Physical Society•Mar 15, 2021•0 citations

Neural-network wavefunctions

Kaitlin McCardle

Nature Computational Science•May 17, 2022•2 citations

Neural Pfaffians: Solving Many Many-Electron Schrödinger Equations

Nicholas Gao, Stephan Günnemann

Neural wave functions accomplished unprecedented accuracies in approximating the ground state of many-electron systems, though at a high computational cost. Recent works proposed amortizing the cost by learning generalized wave functions across different structures and compounds instead of solving each problem independently. Enforcing the permutation antisymmetry of electrons in such generalized neural wave functions remained challenging as existing methods require discrete orbital selection via non-learnable hand-crafted algorithms. This work tackles the problem by defining overparametrized, fully learnable neural wave functions suitable for generalization across molecules. We achieve this by relying on Pfaffians rather than Slater determinants. The Pfaffian allows us to enforce the antisymmetry on arbitrary electronic systems without any constraint on electronic spin configurations or molecular structure. Our empirical evaluation finds that a single neural Pfaffian calculates the ground state and ionization energies with chemical accuracy across various systems. On the TinyMol dataset, we outperform the `gold-standard' CCSD(T) CBS reference energies by 1.9m$E_h$ and reduce energy errors compared to previous generalized neural wave functions by up to an order of magnitude.

arXiv:2405.14762•neural information processing systems•Jan 1, 2024•11 citations

Generalisierte Neuronale Wellenfunktionen

Nicholas Gao

Neural network-based variational Monte Carlo methods have recently emerged as high-accuracy solutions to the electronic Schrödinger equation. This thesis introduces generalized neural wave functions - a single model representing the ground states of various molecules. This approach enables fast and high-accuracy electronic structure calculations by amortizing the cost over many structures. We develop generalized neural wave functions in stages from conformers to arbitrary molecules.

•Jan 12, 2026•0 citations

A neural network wave formalism

Willard L. Miranker

Advances in Applied Mathematics•Nov 16, 2005•8 citations
Neural wave functions accomplished unprecedented accuracies in approximating the ground state of many-electron systems, though at a high computational cost. Recent works proposed amortizing the cost by learning generalized wave functions across different structures and compounds instead of solving each problem independently. Enforcing the permutation antisymmetry of electrons in such generalized neural wave functions remained challenging as existing methods require discrete orbital selection via non-learnable hand-crafted algorithms. This work tackles the problem by defining overparametrized, fully learnable neural wave functions suitable for generalization across molecules. We achieve this by relying on Pfaffians rather than Slater determinants. The Pfaffian allows us to enforce the antisymmetry on arbitrary electronic systems without any constraint on electronic spin configurations or molecular structure. Our empirical evaluation finds that a single neural Pfaffian calculates the ground state and ionization energies with chemical accuracy across various systems. On the TinyMol dataset, we outperform the `gold-standard' CCSD(T) CBS reference energies by 1.9mEhE_hEh​ and reduce energy errors compared to previous generalized neural wave functions by up to an order of magnitude.