Understanding and eliminating spurious modes in variational Monte Carlo using collective variables (2211.09767v1)
Abstract: The use of neural network parametrizations to represent the ground state in variational Monte Carlo (VMC) calculations has generated intense interest in recent years. However, as we demonstrate in the context of the periodic Heisenberg spin chain, this approach can produce unreliable wave function approximations. One of the most obvious signs of failure is the occurrence of random, persistent spikes in the energy estimate during training. These energy spikes are caused by regions of configuration space that are over-represented by the wave function density, which are called ``spurious modes'' in the machine learning literature. After exploring these spurious modes in detail, we demonstrate that a collective-variable-based penalization yields a substantially more robust training procedure, preventing the formation of spurious modes and improving the accuracy of energy estimates. Because the penalization scheme is cheap to implement and is not specific to the particular model studied here, it can be extended to other applications of VMC where a reasonable choice of collective variable is available.
- J. Gubernatis, N. Kawashima, and P. Werner, Quantum Monte Carlo Methods: Algorithms for Lattice Models (Cambridge University Press, 2016).
- F. Becca and S. Sorella, Quantum Monte Carlo Approaches for Correlated Systems (Cambridge University Press, 2017).
- G. Carleo and M. Troyer, Solving the quantum many-body problem with artificial neural networks, Science 355, 602 (2017).
- D. Luo and B. K. Clark, Backflow transformations via neural networks for quantum many-body wave functions, Phys. Rev. Lett. 122, 226401 (2019).
- J. Hermann, Z. Schätzle, and F. Noé, Deep-neural-network solution of the electronic Schrödinger equation, Nature Chemistry 12, 891 (2020).
- G. Cybenko, Approximation by superpositions of a sigmoidal function, Mathematics of Control, Signals and Systems 2, 303 (1989).
- L. Yang, W. Hu, and L. Li, Scalable variational Monte Carlo with graph neural ansatz (2020).
- C.-Y. Park and M. J. Kastoryano, Geometry of learning neural quantum states, Phys. Rev. Research 2, 023232 (2020).
- I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning (MIT Press, 2016) http://www.deeplearningbook.org.
- R. H. Swendsen and J.-S. Wang, Replica Monte Carlo simulation of spin-glasses, Phys. Rev. Lett. 57, 2607 (1986).
- E. M. Boczko and C. L. Brooks, First-principles calculation of the folding free energy of a three-helix bundle protein, Science 269, 393 (1995).
- C. Pangali, M. Rao, and B. J. Berne, A Monte Carlo simulation of the hydrophobic interaction, The Journal of Chemical Physics 71, 2975 (1979).
- S. Bernèche and B. Roux, Energetics of ion conduction through the k+ channel, Nature 414, 73 (2001).
- R. J. Webber and M. Lindsey, Rayleigh-Gauss-Newton optimization with enhanced sampling for variational Monte Carlo, Phys. Rev. Research 4, 033099 (2022).
- H. Bethe, Zur theorie der metalle, Zeitschrift für Physik 71, 205 (1931).
- J. Kästner, Umbrella sampling, WIREs Computational Molecular Science 1, 932 (2011).
- W. Wirtinger, Zur formalen theorie der funktionen von mehr komplexen veränderlichen, Mathematische Annalen 97, 357 (1927).
- C. Predescu, M. Predescu, and C. V. Ciobanu, On the efficiency of exchange in parallel tempering monte carlo simulations, The Journal of Physical Chemistry B 109, 4189 (2005).