Zero-Error Classical Channel Capacity and Simulation Cost Assisted by Quantum Non-Signalling Correlations.

Chengkai Zhu, Xuanqiang Zhao, Xin Wang

Ching–Yi Lai, Runyao Duan

Duan and Winter studied the one-shot zero-error classical capacity of a quantum channel assisted by quantum non-signalling correlations, and formulated this problem as a semidefinite program depending only on the Kraus operator space of the channel. For the class of classical-quantum channels, they showed that the asymptotic zero-error classical capacity assisted by quantum non-signalling correlations, minimized over all classicalquantum channels with a confusability graph G, is exactly log ϑ(G), where ϑ(G) is the celebrated Lov´asz theta function. In this paper, we show that the one-shot capacity for a classical-quantum channel, induced from a circulant graph G defined by equal-sized cyclotomic cosets, is logbϑ(G)c, which further implies that its asymptotic capacity is log ϑ(G). This type of graphs include the cycle graphs of odd length, the Paley graphs of prime vertices, and the cubit residue graphs of prime vertices. Examples of other graphs are also discussed. This gives Lov´asz ϑ function another operational meaning in zero-error classical-quantum communication.

Elloá B. Guedes, Francisco de Assis, Rex A. C. Medeiros

Runyao Duan, Andreas Winter

We study the one-shot zero-error classical capacity of a quantum channel assisted by quantum no-signalling correlations, and the reverse problem of exact simulation of a prescribed channel by a noiseless classical one. Quantum no-signalling correlations are viewed as two-input and two-output completely positive and trace preserving maps with linear constraints enforcing that the device cannot signal. Both problems lead to simple semidefinite programmes (SDPs) that depend only on the Choi-Kraus (operator) space of the channel. In particular, we show that the zero-error classical simulation cost is precisely the conditional min-entropy of the Choi-Jamiołkowski matrix of the given channel. The zero-error classical capacity is given by a similar-looking but different SDP; the asymptotic zero-error classical capacity is the regularization of this SDP, and in general, we do not know of any simple form. Interestingly, however, for the class of classical-quantum channels, we show that the asymptotic capacity is given by a much simpler SDP, which coincides with a semidefinite generalization of the fractional packing number suggested earlier by Aram Harrow. This finally results in an operational interpretation of the celebrated Lovász ϑ function of a graph as the zero-error classical capacity of the graph assisted by quantum no-signalling correlations, the first information theoretic interpretation of the Lovász number.

Xin Wang, Runyao Duan

Using one channel to simulate another exactly with the aid of quantum no-signalling correlations has been studied recently. The one-shot no-signalling assisted classical zero-error simulation cost of non-commutative bipartite graphs has been formulated as semidefinite programms [Duan and Winter, IEEE Trans. Inf. Theory 62, 891 (2016)]. Before our work, it was unknown whether the one-shot (or asymptotic) no-signalling assisted zero-error classical simulation cost for general non-commutative graphs is multiplicative (resp. additive) or not. In this paper we address these issues and give a general sufficient condition for the multiplicativity of the one-shot simulation cost and the additivity of the asymptotic simulation cost of non-commutative bipartite graphs, which include all known cases such as extremal graphs and classical-quantum graphs. Applying this condition, we exhibit a large class of so-called cheapest-full-rank graphs whose asymptotic zero-error simulation cost is given by the one-shot simulation cost. Finally, we disprove the multiplicativity of one-shot simulation cost by explicitly constructing a special class of qubit-qutrit non-commutative bipartite graphs.

Xin Li, Yinan Li

We study zero-error communication and exact simulation of quantum channels assisted by quantum no-signalling correlations. For a channel with Kraus operator space $K$, we introduce a semidefinite programming parameter $Γ(K)$ that is exactly multiplicative under tensor products and upper-bounds $Θ_{\rm NS}(K)$, where $\log_2Θ_{\rm NS}(K)=C_{0,\rm NS}(K)$ and $C_{0,\rm NS}(K)$ denotes the asymptotic quantum no-signalling-assisted zero-error capacity. Among known upper bounds on $Θ_{\rm NS}$ given by semidefinite programs, $Γ$ is the tightest. We also prove that the previously introduced simulation lower bound $Σ^{-}(K)$ is exactly multiplicative and satisfies $Θ_{\rm NS}(K)\leqΓ(K)\leqΣ^{-}(K)\leqΓ(K)^2$. Using this multiplicativity, we show that neither $S_{0,\rm NS}(K)=\log_2Σ^{-}(K)$ nor its regularized extension $S_{0,\rm NS}(K)=\lim\nolimits_{n\to\infty}n^{-1}\log_2Σ^{-}(K^{\otimes n})$ holds in general. This answers an open question about the asymptotic simulation cost $S_{0,\rm NS}(K)$. For classical-quantum channels, we obtain sharp bounds. If $r$ is the effective output dimension, then $Θ_{\rm NS}\leqΓ\leq(1+r^{1/2})Θ_{\rm NS}/2$ and $Γ\leqΣ^{-}\leq(1+r^{1/2})Γ/2$. If $k$ is the minimum rank of the output support projections, then $Σ^{-}/Θ_{\rm NS}\leq[(1+r^{1/2})/(1+k^{1/2})]^2$, and this bound is sharp for every $1\leq k\leq r$. Thus, we obtain the sharp bound $S_{0,\rm NS}-C_{0,\rm NS}\leq2\log_2[(1+r^{1/2})/(1+k^{1/2})]$.