{"id":"arxiv/2608.22925","url":"https://arxiv.org/abs/2608.22925","tags":["physics.chem-ph","cond-mat.str-el","quant-ph"],"score":0,"title":"Importance-Reweighted Fock-Space Variational Monte Carlo","domain":"physics.chem-ph","source":"arxiv","authors":["Zheng Che"],"reviews":0,"subject":"physics.chem-ph","summary":"Fock-space variational Monte Carlo (FS-VMC) evaluates variational quantities by sampling discrete many-body configurations. In molecular applications, concentrated Born distributions can hinder Markov-chain mixing, while Monte Carlo estimators can also exhibit large variance. We introduce importance-reweighted FS-VMC (IR-FS-VMC), which combines Metropolis--Hastings sampling, an evaluation kernel, and self-normalized importance sampling to estimate the Born-distribution energy, gradient, and stochastic reconfiguration (SR) matrix. Controlled comparisons on equilibrium H2O and Fe2S2 show that…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-08-24","reactions":{}}