#### Room 117

#### Device-Independent Tests of Quantum Measurements

#### Sibasish Ghosh

##### IMSc

*Given a device which is `committed' to perform a specific measurement on a given quantum mechanical system, how does one guarantee that the device can actually perform the job? This issue arises every time an experimentalist wants to calibrate a measurement device before using it for the system. By looking at the output probability distributions corresponding to performing measurement with the said device on all possible states of the system, one can, in principle, identify the device. And, in particular, if the (convex) set of all such output probability distributions does have a concrete geometrical structure, it should be somewhat easy to identify the measuring device. Michele Dall' Arno et al. [PRL, vol. 118, 250501 (2017)] addressed this problem and provided analytical solution to it for every qubit POVM. I will discuss about their approach and (if possible) a few open questions.*

Paper link: journals.aps.org/prl/pdf/10.1103/PhysRevLett.118.250501

Done