We give an operational meaning to the min-entropy of a quantum state as a resource measure for various interconnected tasks. In particular, we show that the min-entropy without smoothing measures the amount of quantum information that can be hidden or encoded perfectly in the one-shot setting when the quantum state is used as a randomness or correlation source. First, we show that the min-entropy of entanglement of a pure bipartite state is the maximum number of qubits privately transferable when the state is used as a quantum one-time pad. Then, through the equivalence of quantum secret sharing--like protocols, it is also shown that the min-entropy of a quantum state is the maximum number of qubits that can be masked when the state is used as a randomness source for a quantum masking process. Consequently, we show that the min-entropy of a quantum state is half the size of the quantum state it can catalytically dephase. This gives a necessary and sufficient condition for catalysts for state transition processes.
We give an operational meaning to the min-entropy of a quantum state as a resource measure for various interconnected tasks. In particular, we show that the min-entropy without smoothing measures the amount of quantum information that can be hidden or encoded perfectly in the one-shot setting when the quantum state is used as a randomness or correlation source. First, we show that the min-entropy of entanglement of a pure bipartite state is the maximum number of qubits privately transferable when the state is used as a quantum one-time pad. Then, through the equivalence of quantum secret sharing--like protocols, it is also shown that the min-entropy of a quantum state is the maximum number of qubits that can be masked when the state is used as a randomness source for a quantum masking process. Consequently, we show that the min-entropy of a quantum state is half the size of the quantum state it can catalytically dephase. This gives a necessary and sufficient condition for catalysts for state transition processes.
Catalysts are substances that assist transformation of other resourceful objects without being consumed in the process. However, the fact that their ``catalytic power'' is limited and can be depleted is often overlooked, especially in the recently developing theories on catalysis of quantum randomness utilizing building correlation with catalyst. In this work, we establish a resource theory of one-shot catalytic randomness in which uncorrelatedness is consumed in catalysis of randomness. We do so by completely characterizing bipartite unitary operators that can be used to implement catalysis of randomness using partial transpose. By doing so, we find that every catalytic channel is factorizable, and therefore there exists a unital channel that is not catalytic. We define a family of catalytic entropies that quantifies catalytically extractable R\'enyi entropies from a quantum state and show how much the degeneracy of a quantum state can boost the catalytic entropy beyond its ordinary entropy. Based on this, we demonstrate that a randomness source can be actually exhausted after a certain amount of randomness is extracted. We apply this theory to systems under superselection rules that forbids superposition of certain quantum states and find that nonmaximally mixed states can yield the maximal catalytic entropy. We discuss implications of this theory to various topics, including catalytic randomness absorption, the no-secret theorem, and the possibility of multiparty infinite catalysis.
In quantum Shannon theory, various kinds of quantum entropies are used to characterize the capacities of noisy physical systems. Among them, min-entropy and its smooth version attract wide interest especially in the field of quantum cryptography as they can be used to bound the information obtained by an adversary. However, calculating the exact value or nontrivial bounds of min-entropy are extremely difficult because the composite system dimension may scale exponentially with the dimension of its subsystem. Here, we develop a one-shot lower bound calculation technique for the min-entropy of a classical-quantum state that is applicable to both finite and infinite dimensional reduced quantum states. Moreover, we show our technique is of practical interest in at least three situations. First, it offers an alternative tight finite-data analysis for the BB84 quantum key distribution scheme. Second, it gives the best finite-key bound known to date for a variant of device independent quantum key distribution protocol. Third, it provides a security proof for a novel source-independent continuous-variable quantum random number generation protocol. These results show the effectiveness and wide applicability of our approach.
Quantum entanglement of pure states is usually quantified via the entanglement entropy, the von Neumann entropy of the reduced state. Entanglement entropy is closely related to entanglement distillation, a process for converting quantum states into singlets, which can then be used for various quantum technological tasks. The relation between entanglement entropy and entanglement distillation has been known only for the asymptotic setting, and the meaning of entanglement entropy in the single-copy regime has so far remained open. Here we close this gap by considering entanglement catalysis. We prove that entanglement entropy completely characterizes state transformations in the presence of entangled catalysts. Our results imply that entanglement entropy quantifies the amount of entanglement available in a bipartite pure state to be used for quantum information processing, giving asymptotic results an operational meaning also in the single-copy setup.
A natural measure for the amount of quantum information that a physical system E holds about another system A = A1, .. . , Anis given by the min-entropy Hmin(A|E). In particular, the min-entropy measures the amount of entanglement between E and A, and is the relevant measure when analyzing a wide variety of problems ranging from randomness extraction in quantum cryptography, decoupling used in channel coding, to physical processes such as thermalization or the thermodynamic work cost (or gain) of erasing a quantum system. As such, it is a central question to determine the behavior of the minentropy after some process M. is applied to the system A. Here, we introduce a new generic tool relating the resulting min-entropy to the original one, and apply it to several settings of interest. A simple example of such a process is the one of sampling, where a subset S of the systems A1, ... , An is selected at random. Our tool allows us to quantify the entanglement that E has with the selected systems AS, i.e., Hmin(AS|ES) as a function of the original Hmin(A|E). We give two applications of this result. First, it directly provides the first local quantum-to-classical randomness extractors for use in quantum cryptography, as well as decoupling operations acting on only a small fraction AS of the input A. Moreover, it gives lower bounds on the dimension of k-out-of-n fully quantum random access encodings. Another natural example of such a process is a measurement in, e.g., BB84 bases commonly used in quantum cryptography. We establish the first entropic uncertainty relations with quantum side information that are nontrivial whenever E is not maximally entangled with A. As a consequence, we are able to prove optimality of quantum cryptographic schemes in the noisy-storage model. This model allows for the secure implementation of two-party cryptographic primitives under the assumption that the adversary cannot store quantum information perfectly. A special case is the bounded-quantum-storage model (BQSM), which assumes that the adversary's quantum memory device is noise free but limited in size. Ever since the inception of the BQSM, it has been a vexing open question to determine whether the security is possible as long as the adversary can only.
We consider an unknown quantum state shared between two parties, Alice and Bob, and ask how much quantum communication is needed to transfer the full state to Bob. This problem is known as state merging and was introduced in [Horodecki et al., Nature, 436, 673 (2005)]. It has been shown that for free classical communication the minimal number of quantum bits that need to be sent from Alice to Bob is given by the conditional von Neumann entropy. However this result only holds asymptotically (in the sense that Alice and Bob share initially many identical copies of the state) and it was unclear how much quantum communication is necessary to merge a single copy. We show that the minimal amount of quantum communication needed to achieve this single-shot state merging is given by minus the smooth conditional min-entropy of Alice conditioned on the environment. This gives an operational meaning to the smooth conditional min-entropy.
In this paper, we show that the conditional min-entropyHmin(A|B) of a bipartite staterhoABis directly related to the maximum achievable overlap with a maximally entangled state if only local actions on theB-part ofrhoABare allowed. In the special case whereAis classical, this overlap corresponds to the probability of guessingAgivenB. In a similar vein, we connect the conditional max-entropyHmax(A|B) to the maximum fidelity ofrhoABwith a product state that is completely mixed onA. In the case whereAis classical, this corresponds to the security ofAwhen used as a secret key in the presence of an adversary holdingB. Because min- and max-entropies are known to characterize information-processing tasks such as randomness extraction and state merging, our results establish a direct connection between these tasks and basic operational problems. For example, they imply that the (logarithm of the) probability of guessingAgivenBis a lower bound on the number of uniform secret bits that can be extracted fromArelative to an adversary holdingB.