VMC's gradient estimators are generically heavy-tailed (no 3/2 moment for Slater–Jastrow); PS-Clip-VMC, which clips energies and per-sample gradients, is provably convergent under weak moments and stabilizes FermiNet training.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
SCALE and ACE are new convolutional backflow architectures for Neural Quantum States that deliver O(N^3) scaling with high accuracy and over 40x speedup on Hubbard and t-J models up to 32x32 lattices.
Neural quantum states yield Born-Oppenheimer and non-Born-Oppenheimer energies for high-pressure atomic hydrogen that match or beat prior projector Monte Carlo results up to 128 atoms while avoiding symmetry assumptions and mass-scale issues.
citing papers explorer
-
Is Variational Monte Carlo Robust? Sharp Moment Thresholds and Heavy-tailed Stochastic Optimization
VMC's gradient estimators are generically heavy-tailed (no 3/2 moment for Slater–Jastrow); PS-Clip-VMC, which clips energies and per-sample gradients, is provably convergent under weak moments and stabilizes FermiNet training.
-
Pareto Frontier of Neural Quantum States: Scalable, Affordable, and Accurate Convolutional Backflow for Strongly Correlated Lattice Fermions
SCALE and ACE are new convolutional backflow architectures for Neural Quantum States that deliver O(N^3) scaling with high accuracy and over 40x speedup on Hubbard and t-J models up to 32x32 lattices.
-
Neural Wave Functions for High-Pressure Atomic Hydrogen
Neural quantum states yield Born-Oppenheimer and non-Born-Oppenheimer energies for high-pressure atomic hydrogen that match or beat prior projector Monte Carlo results up to 128 atoms while avoiding symmetry assumptions and mass-scale issues.