Meta Variational Monte Carlo

Tianchen Zhao, James Stokes, Oliver Knitter, Brian Chen, Shravan Veerapaneni
arXiv, 2020 4 citations

Abstract

An identification is found between meta-learning and the problem of determining the ground state of a randomly generated Hamiltonian drawn from a known ensemble. A model-agnostic meta-learning approach is proposed to solve the associated learning problem and a preliminary experimental study of random Max-Cut problems indicates that the resulting Meta Variational Monte Carlo accelerates training and improves convergence.

Summary

Connects meta-learning to quantum ground-state problems, proposing Meta Variational Monte Carlo that accelerates training and improves convergence on random Max-Cut instances.

Key Contributions

Cite this paper if your work involves meta-learning for variational Monte Carlo, transferable neural quantum state optimization, or learning optimization strategies across quantum systems.

FAQ

What is Meta Variational Monte Carlo?
Meta VMC uses model-agnostic meta-learning (MAML) to find initializations for neural quantum state parameters that transfer well across an ensemble of randomly generated Hamiltonians, accelerating convergence on new problem instances.
How does this differ from standard VMC?
Standard VMC trains from scratch for each Hamiltonian. Meta VMC learns a shared initialization across an ensemble of problems, so fine-tuning on a new instance requires far fewer optimization steps.
What problems does Meta VMC address?
The paper demonstrates on random Max-Cut problems, but the framework applies to any setting where one must solve many related ground-state problems from a known distribution of Hamiltonians.
Is there a journal version of this paper?
Yes. This arXiv preprint (2020) was later published in Quantum Machine Intelligence (2023) under the title "Meta-Variational Quantum Monte Carlo."
Can meta-learning be applied to quantum chemistry?
In principle, yes. If one has an ensemble of molecular Hamiltonians (e.g., varying geometries), meta-learning could provide transferable initializations for neural quantum state solvers.

Keywords

meta variational Monte Carlo meta-learning variational Monte Carlo transferable neural quantum states learning to optimize VMC meta-learning quantum many-body systems neural quantum state transfer

Related Papers

BibTeX

@article{zhao2020meta_vmc,
  title={Meta Variational Monte Carlo},
  author={Zhao, Tianchen and Stokes, James and Knitter, Oliver and Chen, Brian and Veerapaneni, Shravan},
  journal={arXiv preprint arXiv:2011.10614},
  year={2020}
}

Recommended Citation Contexts