Generalizing Neural Wave Functions
- URL: http://arxiv.org/abs/2302.04168v2
- Date: Wed, 31 May 2023 11:06:28 GMT
- Title: Generalizing Neural Wave Functions
- Authors: Nicholas Gao, Stephan G\"unnemann
- Abstract summary: We present a neural network-based reparametrization method that can adapt neural wave functions to different molecules.
We also propose a size-consistent wave function Ansatz, tailored to jointly solve Schr"odinger equations of different molecules.
In both computational chemistry and machine learning, we are the first to demonstrate that a single wave function can solve the Schr"odinger equation of molecules with different atoms jointly.
- Score: 2.088583843514496
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Recent neural network-based wave functions have achieved state-of-the-art
accuracies in modeling ab-initio ground-state potential energy surface.
However, these networks can only solve different spatial arrangements of the
same set of atoms. To overcome this limitation, we present Graph-learned
orbital embeddings (Globe), a neural network-based reparametrization method
that can adapt neural wave functions to different molecules. Globe learns
representations of local electronic structures that generalize across molecules
via spatial message passing by connecting molecular orbitals to covalent bonds.
Further, we propose a size-consistent wave function Ansatz, the Molecular
orbital network (Moon), tailored to jointly solve Schr\"odinger equations of
different molecules. In our experiments, we find Moon converging in 4.5 times
fewer steps to similar accuracy as previous methods or to lower energies given
the same time. Further, our analysis shows that Moon's energy estimate scales
additively with increased system sizes, unlike previous work where we observe
divergence. In both computational chemistry and machine learning, we are the
first to demonstrate that a single wave function can solve the Schr\"odinger
equation of molecules with different atoms jointly.
Related papers
- Neural Pfaffians: Solving Many Many-Electron Schrödinger Equations [58.130170155147205]
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.
This work tackles the problem by defining overparametrized, fully learnable neural wave functions suitable for generalization across molecules.
arXiv Detail & Related papers (2024-05-23T16:30:51Z) - Modeling Non-Covalent Interatomic Interactions on a Photonic Quantum
Computer [50.24983453990065]
We show that the cQDO model lends itself naturally to simulation on a photonic quantum computer.
We calculate the binding energy curve of diatomic systems by leveraging Xanadu's Strawberry Fields photonics library.
Remarkably, we find that two coupled bosonic QDOs exhibit a stable bond.
arXiv Detail & Related papers (2023-06-14T14:44:12Z) - A hybrid quantum-classical algorithm for multichannel quantum scattering
of atoms and molecules [62.997667081978825]
We propose a hybrid quantum-classical algorithm for solving the Schr"odinger equation for atomic and molecular collisions.
The algorithm is based on the $S$-matrix version of the Kohn variational principle, which computes the fundamental scattering $S$-matrix.
We show how the algorithm could be scaled up to simulate collisions of large polyatomic molecules.
arXiv Detail & Related papers (2023-04-12T18:10:47Z) - A Self-Attention Ansatz for Ab-initio Quantum Chemistry [3.4161707164978137]
We present a novel neural network architecture using self-attention, the Wavefunction Transformer (Psiformer)
We show that the Psiformer can be used as a drop-in replacement for other neural networks, often dramatically improving the accuracy of the calculations.
This demonstrates that self-attention networks can learn complex quantum mechanical correlations between electrons, and are a promising route to reaching unprecedented accuracy in chemical calculations on larger systems.
arXiv Detail & Related papers (2022-11-24T15:38:55Z) - First principles physics-informed neural network for quantum
wavefunctions and eigenvalue surfaces [0.0]
We propose a neural network to discover parametric eigenvalue and eigenfunction surfaces of quantum systems.
We apply our method to solve the hydrogen molecular ion.
arXiv Detail & Related papers (2022-11-08T23:22:42Z) - Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave
Functions [2.61072980439312]
In this work, we combine a Graph Neural Network (GNN) with a neural wave function to simultaneously solve the Schr"odinger equation for multiple geometries via VMC.
Compared to existing state-of-the-art networks, our Potential Energy Surface Network (PESNet) speeds up training for multiple geometries by up to 40 times while matching or surpassing their accuracy.
arXiv Detail & Related papers (2021-10-11T07:58:31Z) - Computing molecular excited states on a D-Wave quantum annealer [52.5289706853773]
We demonstrate the use of a D-Wave quantum annealer for the calculation of excited electronic states of molecular systems.
These simulations play an important role in a number of areas, such as photovoltaics, semiconductor technology and nanoscience.
arXiv Detail & Related papers (2021-07-01T01:02:17Z) - Variational Monte Carlo calculations of $\mathbf{A\leq 4}$ nuclei with
an artificial neural-network correlator ansatz [62.997667081978825]
We introduce a neural-network quantum state ansatz to model the ground-state wave function of light nuclei.
We compute the binding energies and point-nucleon densities of $Aleq 4$ nuclei as emerging from a leading-order pionless effective field theory Hamiltonian.
arXiv Detail & Related papers (2020-07-28T14:52:28Z) - A deep neural network for molecular wave functions in quasi-atomic
minimal basis representation [0.0]
We present an adaptation of the recently proposed SchNet for Orbitals (SchNOrb) deep convolutional neural network model [Nature Commun 10, 5024] for electronic wave functions in an optimised quasi-atomic minimal basis representation.
For five organic molecules ranging from 5 to 13 heavy atoms, the model accurately predicts molecular orbital energies and wavefunctions and provides access to derived properties for chemical bonding analysis.
arXiv Detail & Related papers (2020-05-11T06:55:36Z) - Quantum Simulation of 2D Quantum Chemistry in Optical Lattices [59.89454513692418]
We propose an analog simulator for discrete 2D quantum chemistry models based on cold atoms in optical lattices.
We first analyze how to simulate simple models, like the discrete versions of H and H$+$, using a single fermionic atom.
We then show that a single bosonic atom can mediate an effective Coulomb repulsion between two fermions, leading to the analog of molecular Hydrogen in two dimensions.
arXiv Detail & Related papers (2020-02-21T16:00:36Z)
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.