Quantum measurement schemes aim to surpass the standard quantum limit (essentially partition noise) and strive to reach the quantum limit (precision inversely proportional to number of injected particles). One particularly promising category of quantum measurement schemes employs a feedback mechanism: leading particles are detected with the resultant information used to control the instrument in order to extract progressively more information during passage of the pulse.
Clever quantum feedback schemes have been devised but are restricted to ideal conditions. In general quantum feedback schemes are challenging to design so we decided to adapt machine learning theory to quantum information inputs and employ our theory to devise adaptive-feedback quantum measurement schemes. In particular our approach replaces guesswork in quantum measurement by a logical, fully-automatic, programmable routine. We show that our method yields schemes that outperform the best known adaptive scheme for interferometric phase estimation. Furthermore our approach can be adapted to the real-world case where the instrument would learn through trial and error an effective quantum feedback routine.
(PLEASE NOTE ROOM CHANGE)