Concentration of quantum channels with random Kraus operators via matrix Bernstein inequality

TL;DR

This paper develops a method to generate and analyze random quantum channels using matrix Bernstein inequality, demonstrating concentration phenomena and properties like twirling and expansion without relying on Haar or Gaussian matrices.

Contribution

It introduces a new approach to construct and analyze random quantum channels using unitary t-designs and matrix Bernstein inequality, improving bounds and growth conditions for Kraus operators.

Findings

Quantum channels typically become twirling channels with high probability.

Bounds on Schatten p-norm are valid for 1 ≤ p ≤ 2.

Number of Kraus operators can be reduced by powers of log d and t.

Abstract

In this study, we generate quantum channels with random Kraus operators to typically obtain almost twirling quantum channels and quantum expanders. To prove the concentration phenomena, we use matrix Bernstein's inequality. In this way, our random models do not utilize Haar-distributed unitary matrices or Gaussian matrices. Rather, as in the preceding research, we use unitary $t$-designs to generate mixed tenor-product unitary channels acting on $(\mathbb C^{d})^{\otimes t}$. Although our bounds in Schatten $p$-norm are valid only for $1\leq p \leq 2$, we show that they are typically almost twirling quantum channels with the tail bound proportional to $1/\mathrm{poly}(d^t)$, while such bounds were previously constants. The number of required Kraus operators was also improved by powers of $\log d$ and $t$. Such random quantum channels are also typically quantum expanders, but the number of Kraus operators must grow proportionally to $\log d$ in our case. Finally, a new non-unital model of super-operators generated by bounded and isotropic random Kraus operators was introduced, which can be typically rectified to give almost randomizing quantum channels and quantum expanders.