Neural network backflow for ab-initio quantum chemistry

Abstract

The ground state of second-quantized quantum chemistry Hamiltonians provides access to an important set of chemical properties. Wavefunctions based on ML architectures have shown promise in approximating these ground states in a variety of physical systems. In this work, we show how to achieve state-of-the-art energies for molecular Hamiltonians using the the neural network backflow wave-function. To accomplish this, we optimize this ansatz with a variant of the deterministic optimization scheme based on SCI introduced by [Li, et. al JCTC (2023)] which we find works better than standard MCMC sampling. For the molecules we studied, NNBF gives lower energy states than both CCSD and other neural network quantum states. We systematically explore the role of network size as well as optimization parameters in improving the energy. We find that while the number of hidden layers and determinants play a minor role in improving the energy, there is significant improvements in the energy from increasing the number of hidden units as well as the batch size used in optimization with the batch size playing a more important role.

References in corpus (4)

• Restricted-Boltzmann-Machine Learning for Solving Strongly Correlated Quantum Systems
• Variational Quantum Monte Carlo Method with a Neural-Network Ansatz for Open Quantum Systems
• A Non-stochastic Optimization Algorithm for Neural-network Quantum States
• Unifying view of fermionic neural network quantum states: From neural network backflow to hidden fermion determinant states

Cited by in corpus (5)

• Improved Optimization for the Neural-network Quantum States and Tests on the Chromium Dimer
• Simple Fermionic backflow states via a systematically improvable tensor decomposition
• Modal Backflow Neural Quantum States for Anharmonic Vibrational Calculations
• Neuralized Fermionic Tensor Networks for Quantum Many-Body Systems
• Precise Quantum Chemistry calculations with few Slater Determinants