{"id":"arxiv/2604.18506","url":"https://arxiv.org/abs/2604.18506","tags":["quant-ph","physics.comp-ph"],"score":0,"title":"Physics-Informed Neural Networks for Maximizing Quantum Fisher Information in Time-Dependent Many-Body Systems","domain":"quant-ph","source":"arxiv","authors":["Antonio Ferrer-Sánchez","Yolanda Vives-Gilabert","Yue Ban","Xi Chen","José D. Martín-Guerrero"],"reviews":0,"subject":"quant-ph","summary":"Quantum Fisher Information (QFI) sets the ultimate precision limit for parameter estimation and is therefore a central quantity in quantum metrology. In time-dependent many-body systems, however, maximizing QFI is a highly non-trivial task due to the combined effects of non-commutativity, control complexity, and the exponential growth of the Hilbert space. In this work, we present a physics-informed neural network (PINN) framework to address this problem through the learning of counter-diabatic quantum dynamics. Our approach combines a variational PINN formulation with a Magnus-expansion…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-04-20","reactions":{}}