Accurate Computation of Quantum Excited States with Neural Networks

Abstract

We present a variational Monte Carlo algorithm for estimating the lowest excited states of a quantum system which is a natural generalization of the estimation of ground states. The method has no free parameters and requires no explicit orthogonalization of the different states, instead transforming the problem of finding excited states of a given system into that of finding the ground state of an expanded system. Expected values of arbitrary observables can be calculated, including off-diagonal expectations between different states such as the transition dipole moment. Although the method is entirely general, it works particularly well in conjunction with recent work on using neural networks as variational Ansätze for many-electron systems, and we show that by combining this method with the FermiNet and Psiformer Ansätze we can accurately recover vertical excitation energies and oscillator strengths on a range of molecules. Our method is the first deep learning approach to achieve accurate vertical excitation energies, including challenging double excitations, on benzene-scale molecules. Beyond the chemistry examples here, we expect this technique will be of great interest for applications to atomic, nuclear and condensed matter physics.

arXiv commentThe content of this version is identical to the manuscript draft which was sent to Science on May 17th 2024

References in corpus (36)

• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Quantum Monte Carlo methods for nuclear physics
• Deep neural network solution of the electronic Schrödinger equation
• Ab-Initio Solution of the Many-Electron Schrödinger Equation with Deep Neural Networks
• Machine learning for electronically excited states of molecules
• A Mountaineering Strategy to Excited States: Highly-Accurate Reference Energies and Benchmarks
• Optimization of quantum Monte Carlo wave functions by energy minimization
• Green Function Monte Carlo with Stochastic Reconfiguration
• Deep autoregressive models for the efficient variational simulation of many-body quantum systems
• Symmetries and many-body excited states with neural-network quantum states
• A Mountaineering Strategy to Excited States: Highly-Accurate Energies and Benchmarks for Medium Size Molecules
• Reference Energies for Double Excitations
• Backflow Transformations via Neural Networks for Quantum Many-Body Wave-Functions
• Ab initio computations of molecular systems by the auxiliary-field quantum Monte Carlo method
• QUESTDB: a database of highly-accurate excitation energies for the electronic structure community
• Troubleshooting Time-Dependent Density-Functional Theory for Photochemical Applications: Oxirane
• Excited states using semistochastic heat-bath configuration interaction
• Reference Energies for Intramolecular Charge-Transfer Excitations
• Ab-initio quantum chemistry with neural-network wavefunctions
• Pfaffian pairing and backflow wave functions for electronic structure quantum Monte Carlo methods
• A Mountaineering Strategy to Excited States: Highly-Accurate Energies and Benchmarks for Exotic Molecules and Radicals
• Optimization of ground and excited state wavefunctions and van der Waals clusters
• Optimized Jastrow-Slater wave functions for ground and excited states: Application to the lowest states of ethene
• Targeted Excited State Algorithms
• New generation of effective core potentials from correlated calculations: 2nd row elements
• A Mountaineering Strategy to Excited States: Highly-Accurate Oscillator Strengths and Dipole Moments of Small Molecules
• Ab initio calculation of real solids via neural network ansatz
• Phases of two-dimensional spinless lattice fermions with first-quantized deep neural-network quantum states
• Quantum Monte Carlo calculations of energy gaps from first principles
• Message-Passing Neural Quantum States for the Homogeneous Electron Gas
• Excited states in variational Monte Carlo using a penalty method
• A Mountaineering Strategy to Excited States: Revising Reference Values with EOM-CC4
• Deep-neural-network approach to solving the ab initio nuclear structure problem
• Tailoring CIPSI expansions for QMC calculations of electronic excitations: the case study of thiophene
• Variational Excitations in Real Solids: Optical Gaps and Insights into Many-Body Perturbation Theory
• Optical Spectroscopy of Comet C/2014 Q2 (Lovejoy) from MIRO

Cited by in corpus (12)

• The QUEST Database of Highly-Accurate Excitation Energies
• Spiers Memorial Lecture: How to do impactful research in artificial intelligence for chemistry and materials science
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• Analyzing Atomic Interactions in Molecules as Learned by Neural Networks
• Is attention all you need to solve the correlated electron problem?
• Machine learning of phases and structures for model systems in physics
• Deep quantum Monte Carlo approach for polaritonic chemistry
• Physics-informed Transformers for Electronic Quantum States
• Bound on entanglement in neural quantum states
• Modal Backflow Neural Quantum States for Anharmonic Vibrational Calculations
• Machine learning the single-$\Lambda$ hypernuclei with neural-network quantum states
• Imaginary Time Spectral Transforms for Excited State Preparation