Using post-measurement information in state discrimination

TL;DR

This paper investigates how post-measurement information can enhance quantum state discrimination, providing optimality conditions, bounds, and highlighting fundamental differences between classical and quantum scenarios.

Contribution

It introduces optimality conditions for measurements with post-measurement information and demonstrates how quantum advantages can be characterized and bounded.

Findings

Optimality conditions for measurements with post-measurement information

Bounds on success probabilities in quantum state discrimination

Quantum post-measurement information can outperform classical scenarios

Abstract

We consider a special form of state discrimination in which after the measurement we are given additional information that may help us identify the state. This task plays a central role in the analysis of quantum cryptographic protocols in the noisy-storage model, where the identity of the state corresponds to a certain bit string, and the additional information is typically a choice of encoding that is initially unknown to the cheating party. We first provide simple optimality conditions for measurements for any such problem, and show upper and lower bounds on the success probability. For a certain class of problems, we furthermore provide tight bounds on how useful post-measurement information can be. In particular, we show that for this class finding the optimal measurement for the task of state discrimination with post-measurement information does in fact reduce to solving a different problem of state discrimination without such information. However, we show that for the corresponding classical state discrimination problems with post-measurement information such a reduction is impossible, by relating the success probability to the violation of Bell inequalities. This suggests the usefulness of post-measurement information as another feature that distinguishes the classical from a quantum world.