Dresden 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
QI: Fachverband Quanteninformation
QI 20: Metrology and Sensing
QI 20.5: Vortrag
Freitag, 13. März 2026, 10:30–10:45, BEY/0137
Iterative optimization in quantum metrology and entanglement theory using semidefinite programming — Árpád Lukács1,2,3, •Róbert Trényi2,3,4,5, Tamás Vértesi6, and Géza Tóth3,4,2 — 1Dept. of Math. Sci., Durh. Univ., UK — 2HUN-REN Wigner RCP, Budapest, Hungary — 3Dept. of Th. Phys., UPV/EHU, Bilbao, Spain — 4EHU Quantum Center, UPV/EHU, Bilbao, Spain — 5Dept. of Th. Phys., Univ. of Szeged, Hungary — 6HUN-REN Inst. for Nucl. Research, Debrecen, Hungary
Metrological performance of a quantum state is measured by how much it can outperform all separable states in a metrological task. We present efficient optimization techniques to maximize this performance by searching for the optimal local Hamiltonian generating the unitary dynamics for a given bipartite initial state. We show that this is equivalent to maximizing the Quantum Fisher Information over a specific set of local Hamiltonians. This task is very difficult, as it involves maximizing a convex function over a convex set. We reformulate the problem in a bilinear way and optimize it using an iterative see-saw method, where each optimization step is solved via semidefinite programming. We further show that the same optimization framework can be adapted to problems in entanglement theory, such as identifying bound entangled states that maximally violate the Computable Cross Norm-Realignment criterion. Finally, we provide examples where two copies of a quantum state outperform a single copy, demonstrating metrological activation for certain small systems.
Keywords: quantum metrology; entanglement theory; iterative see-saw method; quantum fisher information