Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Leibniz international proceedings in informatics

Quantum Enhancement of Randomness Distribution

Raúl García−Patrón, William Matthews, Andreas Winter

Published Jan 1, 2015• 1 citation
Publisher

Abstract

The capability of a given channel to transmit information is, a priori, distinct from its capability to distribute random correlations. Despite that, for classical channels, the capacity to distribute information and randomness turns out to be the same, even with the assistance of auxiliary communication. In this work we show that this is no longer true for quantum channels when feedback is allowed. We prove this by constructing a channel that is noisy for the transmission of information but behaves as a virtual noiseless channel for randomness distribution when assisted by feedback communication. Our result can be seen as a way of unlocking quantum randomness internal to the channel.

DOI
10.4230/lipics.tqc.2015.180

Related papers

Quantum Enhancement of Randomness Distribution

Raúl García−Patrón, William Matthews, Andreas Winter

The capability of a given channel to communicate information is, a priori, distinct from its capability to distribute shared randomness. In this paper, we define randomness distribution capacities of quantum channels assisted by forward, back, or two-way classical communication and compare these to the corresponding communication capacities. With forward assistance or no assistance, we find that they are equal. We establish the mutual information of the channel as an upper bound on the two-way assisted randomness distribution capacity. This implies that all of the capacities are equal for classical-quantum channels. On the other hand, we show that the back-assisted randomness distribution capacity of a quantum-classical channel is equal to its mutual information. This is often strictly greater than the back-assisted communication capacity. We give an explicit example of such a separation where the randomness distribution protocol is noiseless.

arXiv:1610.05190•IEEE Transactions on Information Theory•Mar 30, 2018•4 citations

Super-Activation of Zero-Error Capacity of Noisy Quantum Channels

Runyao Duan

We study various super-activation effects in the following zero-error communication scenario: One sender wants to send classical or quantum information through a noisy quantum channel to one receiver with zero probability of error. First we show that there are quantum channels of which a single use is not able to transmit classical information perfectly yet two uses can. This is achieved by employing entangled input states between different uses of the given channel and thus cannot happen for classical channels. Second we exhibit a class of quantum channel with vanishing zero-error classical capacity such that when a noiseless qubit channel or one ebit shared entanglement are available, it can be used to transmit $\log_2 d$ noiseless qubits, where 2d is the dimension of input state space. Third we further construct quantum channels with vanishing zero-error classical capacity when assisted with classical feedback can be used to transmit both classical and quantum information perfectly. These striking findings not only indicate both the zero-error quantum and classical capacities of quantum channels satisfy a strong super-additivity beyond any classical channels, but also highlight the activation power of auxiliary physical resources in zero-error communication.

arXiv:0906.2527•Jun 15, 2009•96 citations

Classical feedback for quantum channels

Andrew Skeen

We investigate whether the use of a noiseless, classical feedback channel will increase the capacity of a quantum discrete memoryless channel to transmit classical information. This problem has been previously analyzed by Bowen and Nagarajan for the case of protocols restricted to product input states. They showed that feedback did not increase the information capacity. In this paper we introduce a quantum analogue of classical causality and prove a capacity theorem (in regularized form) for the transmission of classical information.

arXiv:quant-ph/0409207•Sep 29, 2004•0 citations

Inequalities and Separations Among Assisted Capacities of Quantum Channels

Charles H. Bennett, Igor Devetak, Peter W. Shor +1 more

We exhibit quantum channels whose classical and quantum capacities, when assisted by classical feedback, exceed their unassisted classical Holevo capacity. These channels are designed to be noisy in a way that can be corrected with the help of the output and a reference system entangled with part of the input. A similar construction yields quantum channels whose classical capacity, when assisted by two-way classical communication independent of the source, exceeds their classical capacity assisted by feedback alone. We give a hierarchy of capacity inequalities and open questions.

arXiv:quant-ph/0406086•Physical Review Letters•Apr 18, 2006•46 citations

Quantum Feedback Channels

Garry Bowen

In Shannon information theory, the capacity of a memoryless communication channel cannot be increased by the use of feedback. In quantum information theory, the no-cloning theorem means that noiseless copying and feedback of quantum information cannot be achieved. In this correspondence, quantum feedback is defined as the unlimited use of a noiseless quantum channel from receiver to sender. Given such quantum feedback, it is shown to provide no increase in the entanglement-assisted capacities of a memoryless quantum channel, in direct analogy to the classical case. It is also shown that in various cases of nonassisted capacities, feedback may increase the capacity of memoryless quantum channels.

arXiv:quant-ph/0209076•IEEE Transactions on Information Theory•Sep 28, 2004•50 citations

On Feedback and the Classical Capacity of a Noisy Quantum Channel

Garry Bowen, Rajagopal Nagarajan

In Shannon information theory, the capacity of a memoryless communication channel cannot be increased by the use of feedback from receiver to sender. In this correspondence, the use of classical feedback is shown to provide no increase in the unassisted classical capacity of a memoryless quantum channel when feedback is used across nonentangled input states, or when the channel is an entanglement-breaking channel. This gives a generalization of the Shannon theory for certain classes of feedback protocols when transmitting through noisy quantum communication channels.

arXiv:quant-ph/0305176•IEEE Transactions on Information Theory•Jan 1, 2005•27 citations
The capability of a given channel to transmit information is, a priori, distinct from its capability to distribute random correlations. Despite that, for classical channels, the capacity to distribute information and randomness turns out to be the same, even with the assistance of auxiliary communication. In this work we show that this is no longer true for quantum channels when feedback is allowed. We prove this by constructing a channel that is noisy for the transmission of information but behaves as a virtual noiseless channel for randomness distribution when assisted by feedback communication. Our result can be seen as a way of unlocking quantum randomness internal to the channel.
The capability of a given channel to communicate information is, a priori, distinct from its capability to distribute shared randomness. In this paper, we define randomness distribution capacities of quantum channels assisted by forward, back, or two-way classical communication and compare these to the corresponding communication capacities. With forward assistance or no assistance, we find that they are equal. We establish the mutual information of the channel as an upper bound on the two-way assisted randomness distribution capacity. This implies that all of the capacities are equal for classical-quantum channels. On the other hand, we show that the back-assisted randomness distribution capacity of a quantum-classical channel is equal to its mutual information. This is often strictly greater than the back-assisted communication capacity. We give an explicit example of such a separation where the randomness distribution protocol is noiseless.
We study various super-activation effects in the following zero-error communication scenario: One sender wants to send classical or quantum information through a noisy quantum channel to one receiver with zero probability of error. First we show that there are quantum channels of which a single use is not able to transmit classical information perfectly yet two uses can. This is achieved by employing entangled input states between different uses of the given channel and thus cannot happen for classical channels. Second we exhibit a class of quantum channel with vanishing zero-error classical capacity such that when a noiseless qubit channel or one ebit shared entanglement are available, it can be used to transmit log⁡2d\log_2 dlog2​d noiseless qubits, where 2d is the dimension of input state space. Third we further construct quantum channels with vanishing zero-error classical capacity when assisted with classical feedback can be used to transmit both classical and quantum information perfectly. These striking findings not only indicate both the zero-error quantum and classical capacities of quantum channels satisfy a strong super-additivity beyond any classical channels, but also highlight the activation power of auxiliary physical resources in zero-error communication.
We investigate whether the use of a noiseless, classical feedback channel will increase the capacity of a quantum discrete memoryless channel to transmit classical information. This problem has been previously analyzed by Bowen and Nagarajan for the case of protocols restricted to product input states. They showed that feedback did not increase the information capacity. In this paper we introduce a quantum analogue of classical causality and prove a capacity theorem (in regularized form) for the transmission of classical information.
We exhibit quantum channels whose classical and quantum capacities, when assisted by classical feedback, exceed their unassisted classical Holevo capacity. These channels are designed to be noisy in a way that can be corrected with the help of the output and a reference system entangled with part of the input. A similar construction yields quantum channels whose classical capacity, when assisted by two-way classical communication independent of the source, exceeds their classical capacity assisted by feedback alone. We give a hierarchy of capacity inequalities and open questions.
In Shannon information theory, the capacity of a memoryless communication channel cannot be increased by the use of feedback. In quantum information theory, the no-cloning theorem means that noiseless copying and feedback of quantum information cannot be achieved. In this correspondence, quantum feedback is defined as the unlimited use of a noiseless quantum channel from receiver to sender. Given such quantum feedback, it is shown to provide no increase in the entanglement-assisted capacities of a memoryless quantum channel, in direct analogy to the classical case. It is also shown that in various cases of nonassisted capacities, feedback may increase the capacity of memoryless quantum channels.
In Shannon information theory, the capacity of a memoryless communication channel cannot be increased by the use of feedback from receiver to sender. In this correspondence, the use of classical feedback is shown to provide no increase in the unassisted classical capacity of a memoryless quantum channel when feedback is used across nonentangled input states, or when the channel is an entanglement-breaking channel. This gives a generalization of the Shannon theory for certain classes of feedback protocols when transmitting through noisy quantum communication channels.