Uniform Additivity of Tripartite Optimized Correlation Measures 4 Authors contributed equally
Joshua Levin 1 4 , Ariel Shlosberg 1 2 4 , Vikesh Siddhu 1 , and Graeme Smith 1 3 · 2024
Read Uniform Additivity of Tripartite Optimized Correlation Measures 4 Authors contributed equally by Joshua Levin 1 4 , Ariel Shlosberg 1 2 4 , Vikesh Siddhu 1 , and Graeme Smith 1 3 (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-18586.txt.
Introduction
Joshua Levin, Ariel Shlosberg, Vikesh Siddhu, and Graeme Smith study when sums of entropy terms for three-party quantum states add up cleanly. Their 2024 work serves quantum information theory, where best rates for tasks are often written as limits over many state copies. Such limits can be impossible to compute.
The team hunts linear entropy blends whose values need no such limit. Their method extends a bipartite tool to three parties, using only strong subadditivity of entropy. It yields cone-shaped sets of blends that add uniformly. With those cones they prove three known tripartite measures are additive after all. The result turns hard regularized formulas into tractable single-copy ones.
Catalog record
| CREATOR | Joshua Levin 1 4 , Ariel Shlosberg 1 2 4 , Vikesh Siddhu 1 , and Graeme Smith 1 3 |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 80 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-18586
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Uniform Additivity of Tripartite Optimized Correlation Measures 4 Authors contributed equally Joshua Levin 1 4 , Ariel Shlosberg 1 2 4 , Vikesh Siddhu 1 , and Graeme Smith 1 3 Abstract Information theory provides a framework for answering fundamental questions about the optimal performance of many important quantum communication and computational tasks. In many cases, the optimal rates of these tasks can be expressed in terms of regularized formulas that consist of linear combinations of von Neumann entropies optimized over state extensions. However, evaluation of regularized formulas is often intractable, since it involves computing a formula’s value in the limit of infinitely many copies of a state. To find optimized, linear entropic functions of quantum states whose regularized versions are tractable to compute, we search for linear combinations of entropies on tripartite quantum states that are additive. We use the method of [ 1 ] , which considers bipartite formulas, to identify convex polyhedral cones of uniformly additive tripartite correlation measures. We rely only on strong subadditivity of the von Neumann entropy and use these cones to prove that three previously established tripartite optimized correlation measures are additive. I Introduction Often, fundamental bounds on performance in information-theoretic tasks can be expressed through linear combinations of subsystem entropies. As a simple example, the optimal rate at which a quantum state can be compressed is given by the state’s von Neumann entropy, S(\rho)=-\text{Tr}[\rho\log\rho] [ 2 ] . For other types of tasks, the formulas describing optimal performance are more complicated and may involve optimizations over states of certain systems. For instance, the maximum rate at which classical information can be transmitted over a classical channel, which maps a message set X to a random variable Y , is equal to the maximum value of the mutual information of X and Y , \displaystyle\max_{p(X)}I(X:Y)=\max_{p(X)}~[H(X)+H(Y)-H(XY)], (1) where the message set X is treated as a random variable with an associated distribution (or “state”) p(X) that is optimized over in ( 1 ) [ 3 ] . Here, H(X)=-\sum_{x}p(x)\log p(x) is the Shannon entropy. Furthermore, for many tasks involving a fixed input state \rho , the formula expressing optimal performance involves optimizing a linear combination of entropies over the states of some auxiliary systems (referred to as ancillas), and the evaluation must be done in the limit of many copies of \rho . Consider the task of preparing a large number of copies, n , of a bipartite quantum state, \rho_{AB} , via LOCC operations on an initial finite supply of EPR pairs, subject to the restriction that the amount of communication between A and B is sublinear in n . One natural question to consider is what is the fewest number of EPR pairs per copy of \rho_{AB} for which this task is possible? The answer is given by \displaystyle E_{P}^{\infty}(\rho_{AB})\equiv\lim_{n\to\infty}\frac{1}{n}E_{P}(\rho_{AB}^{\otimes n}), (2) where E_{P} is the entanglement of purification [ 4 ] , \displaystyle E_{P}(\rho_{AB})=\min_{\rho_{ABV}}S(\rho_{AV}). (3) So-called regularized formulas like ( 2 ) are very challenging to compute in cases where the optimized formula being evaluated in the large n limit is non-additive, which is likely the case for E_{P} [ 5 ] . For a function E acting on a quantum state to be additive, we mean that \displaystyle E(\rho_{1}\otimes\rho_{2})=E(\rho_{1})+E(\rho_{2}) (4) for any two states \rho_{1} and \rho_{2} . When E is additive, we have that E^{\infty}\equiv\lim_{n\to\infty}\frac{1}{n}E(\rho^{\otimes n})=E , which means that computing the regularized version of E is no more difficult than computing E itself. In order to identify functions of states that have operational interpretations in terms of information theoretic tasks, and whose regularized evaluation might be tractable, it is useful to identify linear combinations of entropies that, when optimized over ancillas, yield additive functions. A systematic method for identifying such functions was introduced in [ 1 ] , where the authors executed their method for functions of bipartite states 1 1 1 Actually, in [ 1 ] , the authors prepare bipartite states by starting with a state on a single system A and then acting on \rho_{A} with a quantum channel \mathcal{N} , which produces a state \rho_{BE} where B is the channel’s output and E is the channel’s environment. The authors then search for functions E of quantum channels for which E(\mathcal{N}\otimes\mathcal{M})=E(\mathcal{N})+E(\mathcal{M}) . This problem is equivalent to the problem of finding functions E of bipartite states for which E(\rho_{1}\otimes\rho_{2})=E_{P}(\rho_{1})+E_{P}(\rho_{2}) . . In this work, we will employ the same method for functions of tripartite states. In so doing, we are able to prove that three tripartite correlation measures introduced in [ 6 ] – E_{c_{2}} , E_{b_{1}} , and E_{s_{2}} -- are additive 2 2 2 This naming scheme, although it may seem arcane, was established to reflect symmetries displayed by these quantities when evaluated using geometric methods in holographic states in AdS/CFT. See [ 6 ] for details. . Like in [ 1 ] , all instances of additivity established in this work are proven using only strong subadditivity of the von Neumann entropy [ 7 ] . Multipartite correlations in quantum states are much less understood than bipartite correlations, and have been the subject of a great deal of research [ 8 ] . While many open questions remain, it has been repeatedly demonstrated that multipartite entanglement plays an important role in a wide range of areas of study, including metrology [ 9 , 10 ] , many-body physics [ 11 , 12 , 13 ] , nonlocal games [ 14 , 15 ] , quantum secret sharing [ 16 ] , device-independent randomness certification [ 17 ] , and distributed quantum computing [ 18 , 19 ] . In the two-party case, questions about how to characterize entanglement and whether a system is maximally entangled have clear answers [ 8 ] . For bipartite pure states, there is only a single type of entanglement and the entanglement entropy, S(\Tr_{B}[\ket{\psi_{AB}}\bra{\psi_{AB}}]) , quantifies the degree of entanglement present. However, already in the case of a three qubit system, the system can be entangled in two inequivalent ways [ 20 ] , and in the case of four qubits, there are already nine inequivalent entanglement classes [ 21 ] . Therefore, as a result of the richness and usefulness of multipartite entanglement, it is worthwhile to determine the set of tripartite correlation measures that have operational meanings. The remainder of this paper is organized as follows. In Sec. II , we review the necessary mathematical background, including optimized linear entropic formulas, and convex polyhedral cones. In Sec. III , we review the methods and results of [ 1 ] , which introduced the notion of uniform additivity and identified the complete set of uniformly additive bipartite optimized linear entropic formulas. In Sec. IV , we apply the method of [ 1 ] to identify uniformly additive tripartite optimized linear entropic formulas. Finally, in Sec. V , we end with some concluding remarks, and comments on future work. II Preliminaries We will begin by reviewing some useful mathematical tools and definitions. First, we review the definition of optimized linear entropic formulas . We then review the definition and properties of convex polyhedral cones along with the related concept of the dual cone . II-A Optimized Linear