Agnostic Tomography of Stabilizer Product States

Abstract
We define a quantum learning task called agnostic tomography, where given copies of an arbitrary state and a class of quantum states , the goal is to output a succinct description of a state that approximates at least as well as any state in (up to some small error ). This task generalizes ordinary quantum tomography of states in and is more challenging because the learning algorithm must be robust to perturbations of . We give an efficient agnostic tomography algorithm for the class of -qubit stabilizer product states. Assuming has fidelity at least with a stabilizer product state, the algorithm runs in time . This runtime is quasipolynomial in all parameters, and polynomial if is a constant.
View on arXivComments on this paper