Quantum Gaussian filter for exploring ground-state properties

Filter methods realize a projection from superposed quantum state onto target state, which can be efficient if two states have sufficient overlap. Here we propose Gaussian filter (QGF) with the operator being function of system Hamiltonian. A hybrid quantum-classical algorithm feasible on near-term computers is developed, implements as linear combination Hamiltonian evolution at various times. Remarkably, coefficients are determined classically and optimized in postprocessing procedure. Compared to existing algorithms whose given advance, our method more flexible practice under resources help classical computers. We demonstrate for Ising model numeral simulations noises. also an alternative full approach that QGF ancillary continuous-variable mode.