Solving quasiparticle band spectra of real solids using neural-network quantum states

Abstract Establishing a predictive ab initio method for solid systems is one of the fundamental goals in condensed matter physics and computational materials science. The central challenge how to encode highly-complex quantum-many-body wave function compactly. Here, we demonstrate that artificial neural networks, known their overwhelming expressibility context machine learning, are excellent tool first-principles calculations extended periodic materials. We show ground-state energies real solids one-, two-, three-dimensional simulated precisely, reaching chemical accuracy. highlight our work quasiparticle band spectra, which both essential peculiar solid-state systems, can be efficiently extracted with technique designed exploit low-lying energy structure from networks. This opens up path elucidate intriguing complex many-body phenomena systems.