Spoofing of Quantum Channels Enables Low-Rank Projective Simulation

TL;DR

This paper investigates quantum channel discrimination under limited measurement resources, revealing gauge freedoms and providing algorithms to minimize Kraus rank, with implications for quantum simulation and device certification.

Contribution

It characterizes equivalence classes of quantum channels with the same measurement outcomes, introduces a Sinkhorn-like algorithm to find minimal Kraus rank, and analyzes spoofing of Pauli channels.

Findings

Kraus rank can be reduced from d^2 to d for generic systems.

A Sinkhorn-like algorithm finds minimal Kraus rank for outcome marginals.

Numerical demonstrations for systems up to dimension 20.

Abstract

The ability to characterise and discern quantum channels is a crucial aspect of noisy quantum technologies. In this work, we explore the problem of distinguishing quantum channels when limited to sub-exponential resources, framed as von Neumann (projective) measurements. We completely characterise equivalence classes of quantum channels with different Kraus ranks that have the same marginal distributions under compatible projective measurements. In doing so, we explicitly identify gauge freedoms which can be varied without changing those compatible marginal outcome distributions, opening new avenues for quantum channel simulation, variational quantum channels, as well as novel adversarial strategies in noisy quantum device certification. Specifically, we show how a Sinkhorn-like algorithm enables us to find the minimum admissible Kraus rank that generates the correct outcome marginals. For a generic $d$-dimensional quantum system, this lowers the Kraus rank from $d^2$ to the theoretical minimum of $d$. For up to $d = 20$, we numerically demonstrate our findings, for which the code is available and open source. Finally, we provide an analytic algorithm for the special case of spoofing Pauli channels.