Search

Based on the linearity of quantum unitary operations, we propose a method that runs the parameterized quantum circuits before encoding the input data. This enables a dataset owner to train machine learning models on quantum cloud computation platforms, without the risk of leaking the information about the data. It is also capable of encoding a vast amount of data effectively at a later time using classical computations, thus saving runtime on quantum computation devices. The trained quantum machine learning models can be run completely on classical computers, meaning the dataset owner does not need to have any quantum hardware, nor even quantum simulators. Moreover, our method mitigates the encoding bottleneck by reducing the required circuit depth from $O(2^{n})$ to $O(n)$, and relax the tolerance on the precision of the quantum gates for the encoding. These results demonstrate yet another advantage of quantum and quantum-inspired machine learning models over existing classical neural networks, and broaden the approaches to data security., Comment: Added experimental results and the flow chart of our method

The no-masking theorem (Phys. Rev. Lett. 120, 230501 (2018)) claims that arbitrary quantum states cannot be masked. Based on this result, the authors further suggested that qubit commitment is not possible. Here we show that this connection does not hold in general., Comment: 2 pages

When computing the gradients of a quantum neural network using the parameter-shift rule, the cost function needs to be calculated twice for the gradient with respect to a single adjustable parameter of the network. When the total number of parameters is high, the quantum circuit for the computation has to be adjusted and run for many times. Here we propose an approach to compute all the gradients using a single circuit only, with a much reduced circuit depth and less classical registers. We also demonstrate experimentally, on both real quantum hardware and simulator, that our approach has the advantages that the circuit takes a significantly shorter time to compile than the conventional approach, resulting in a speedup on the total runtime., Comment: Added a suggestion on improving real quantum computers

We give a general proof showing that once irreversible processes are involved, a class of projective measurements is impossible. Applying this no-go result to the Schroedinger's cat paradox implies that if something is claimed to be a real Schroedinger's cat, there will be no measurable difference between it and a trivial classical mixture of ordinary cats in any physically implementable process, otherwise raising the dead will become reality. Other similar macroscopic quantum superpositions cannot be observed either due to the lack of non-commuting measurement bases. Our proof does not involve any quantum interpretation theory and hypothesis., Comment: 7 pages

We find experimentally that when artificial neural networks are connected in parallel and trained together, they display the following properties. (i) When the parallel-connected neural network (PNN) is optimized, each sub-network in the connection is not optimized. (ii) The contribution of an inferior sub-network to the whole PNN can be on par with that of the superior sub-network. (iii) The PNN can output the correct result even when all sub-networks give incorrect results. These properties are unlikely for natural biological sense organs. Therefore, they could serve as a simple yet effective criterion for measuring the bionic level of neural networks. With this criterion, we further show that when serving as the activation function, the ReLU function can make an artificial neural network more bionic than the sigmoid and Tanh functions do., Comment: 9 pages, 8 figures, 3 tables

Measurement-device-independent quantum key distribution (MDI-QKD) provides a method for secret communication whose security does not rely on trusted measurement devices. In all existing MDI-QKD protocols, the participant Charlie has to perform the Bell state measurement or other joint measurements. Here we propose an MDI-QKD protocol which requires individual measurements only. Meanwhile, all operations of the receiver Bob are classical, without the need for preparing and measuring quantum systems. Thus the implementation of the protocol has a lower technical requirement on Bob and Charlie., Comment: 6 pages, 1 figure

Based on quantum counterfactual interaction-free measurement, we propose an implementation scheme for a beam splitter with anomalous reflection and transmission properties that looks impossible at first glance. Our scheme is stationary without requiring switchable mirrors and polarization rotators. Using the scheme for imaging will ensure that the optical radiation received by the object being imaged can be arbitrarily low. Thus it enables applications such as stealthy night vision devices that can work without detectable ambient light, or being used as a hackware against some counterfactual quantum cryptographic protocols., Comment: This paper has been withdrawn by the author, because the proposed scheme does not work

Unconditionally secure quantum bit commitment (QBC) was widely believed to be impossible for more than two decades. But recently, based on an anomalous behavior found in quantum steering, we proposed a QBC protocol which can be unconditionally secure in principle. The protocol requires the use of infinite-dimensional systems, therefore it may seem less feasible in practice. Here we propose a quantum optical method based on Mach-Zehnder interferometer, which gives a very good approximation to such infinite-dimensional systems. Thus, it enables a proof-of-principle experimental implementation of our protocol, which can also serve as a practically secure QBC scheme. Other multi-party cryptographic protocols such as quantum coin tossing can be built upon it too. Our approach also reveals a relationship between infinity and non-locality, which may have an impact on the research of fundamental theories., Comment: Very close to the published version

In a recent paper (Int. J. Quantum Inf. 17 (2019) 1950026), the authors discussed the shortcomings in the security of a quantum private comparison protocol that we previously proposed (Int. J. Quantum Inf. 15 (2017) 1750014). They also proposed a new protocol aimed to avoid these problems. Here we analysis the information leaked in their protocol, and find that it is even less secure than our protocol in certain cases. We further propose an improved version which has the following advantages: (1) no entanglement needed, (2) quantum memory is no longer required, and (3) less information leaked. Therefore, better security and great feasibility are both achieved., Comment: Very close to the published version

Recently, Frauchiger and Renner proposed a Gedankenexperiment, which was claimed to be able to prove that quantum theory cannot consistently describe the use of itself. Here we show that the conclusions of Frauchiger and Renner actually came from their incorrect description of some quantum states. With the correct description there will be no inconsistent results, no matter which quantum interpretation theory is used. Especially, the Copenhagen interpretation can satisfy all the three assumptions (C), (Q), and (S) of Frauchiger and Renner simultaneously, thus it has no problem consistently describing the use of itself., Comment: Published version

Quantum oblivious transfer (QOT) is an essential cryptographic primitive. But unconditionally secure QOT is known to be impossible. Here we propose a practical QOT protocol, which is perfectly secure against dishonest sender without relying on any technological assumption. Meanwhile, it is also secure against dishonest receiver in the absence of long-term quantum memory and complicated collective measurements. The protocol is extremely feasible, as it can be implemented using currently available Mach-Zehnder interferometer, and no quantum memory, collective measurements nor entanglement are needed for honest participants. More importantly, comparing with other practical QOT schemes, our protocol has an unbeatable efficiency since it requires the transmission of a single photon only., Comment: 7 pages, 2 figures

Since unconditionally secure quantum two-party computations are known to be impossible, most existing quantum private comparison (QPC) protocols adopted a third party. Recently, we proposed a QPC protocol which involves two parties only, and showed that although it is not unconditionally secure, it only leaks an extremely small amount of information to the other party. Here we further propose the device-independent version of the protocol, so that it can be more convenient and dependable in practical applications., Comment: Published version. arXiv admin note: text overlap with arXiv:1604.06834

Unconditionally secure quantum bit commitment (QBC) was considered impossible. But the no-go proofs are based on the Hughston-Jozsa-Wootters (HJW) theorem (a.k.a. the Uhlmann theorem). Recently it was found that in high-dimensional systems, there exist some states which can display a chaos effect in quantum steering, so that the attack strategy based on the HJW theorem has to require the capability of discriminating quantum states with very subtle difference, to the extent that is not allowed by the uncertainty principle. With the help of this finding, here we propose a simple QBC protocol which manages to evade the no-go proofs., Comment: 7 pages. Security proof extended. No change to the protocol

Quantum steering means that in some bipartite quantum systems, the local measurements on one side can determine the state of the other side. Here we show that in high-dimensional systems, there exists a specific entangled state which can display a kind of chaos effect when being adopted for steering. That is, a subtle difference in the measurement results on one side can steer the other side into completely orthogonal states. Moreover, by expanding the result to infinite-dimensional systems, we find two sets of states for which, contrary to common belief, even though their density matrices approach being identical, the steering between them is impossible. This property makes them very useful for quantum cryptography., Comment: Published version

To evade the well-known impossibility of unconditionally secure quantum two-party computations, previous quantum private comparison protocols have to adopt a third party. Here we study how far we can go with two parties only. We propose a very feasible and efficient protocol. Intriguingly, although the average amount of information leaked cannot be made arbitrarily small, we find that it never exceeds 14 bits for any length of the bit-string being compared., Comment: 4 pages, 1 figure

Quantum coin flipping (QCF) is an essential primitive for quantum cryptography. Unconditionally secure strong QCF with an arbitrarily small bias was widely believed to be impossible. But basing on a problem which cannot be solved without quantum algorithm, here we propose such a QCF protocol, and show how it manages to evade all existing no-go proofs on QCF., Comment: Very close to the published version

We propose a cheating strategy to a relativistic quantum commitment scheme [Sci Rep 2014;4:6774] which was claimed to be unconditionally secure. It is shown that the sender Alice can cheat successfully with probability 100%, thus disproving the security claim., Comment: arXiv admin note: Title of previous version was: Comment on "Unconditionally secure commitment in position-based quantum cryptography"

Cheat sensitive quantum bit commitment (CSQBC) loosens the security requirement of quantum bit commitment (QBC), so that the existing impossibility proofs of unconditionally secure QBC can be evaded. But here we analyze the common features in all existing CSQBC protocols, and show that in any CSQBC having these features, the receiver can always learn a non-trivial amount of information on the sender's committed bit before it is unveiled, while his cheating can pass the security check with a probability not less than 50%. The sender's cheating is also studied. The optimal CSQBC protocols that can minimize the sum of the cheating probabilities of both parties are found to be trivial, as they are practically useless. We also discuss the possibility of building a fair protocol in which both parties can cheat with equal probabilities., Comment: Published version

In quantum weak oblivious transfer, Alice sends Bob two bits and Bob can learn one of the bits at his choice. It was found that the security of such a protocol is bounded by $2P_{Alice}^{\ast }+P_{Bob}^{\ast }\geq 2$, where $P_{Alice}^{\ast }$ is the probability with which Alice can guess Bob's choice, and $P_{Bob}^{\ast }$ is the probability with which Bob can guess both of Alice's bits\ given that he learns one of the bits with certainty. Here we propose a protocol and show that as long as Alice is restricted to individual measurements, then both $P_{Alice}^{\ast }$ and $P_{Bob}^{\ast }$ can be made arbitrarily close to 1/2, so that maximal violation of the security bound can be reached. Even with some limited collective attacks, the security bound can still be violated. Therefore, although our protocol still cannot break the bound in principle when Alice has unlimited cheating power, it is sufficient for achieving secure quantum weak oblivious transfer in practice., Comment: Published version

Recently there were many proposals on device-independent (DI) quantum key distribution protocol whose security is based on the violation of the Clauser-Horne-Shimony-Holt inequality. However, as a statistical law, a certain extent of fluctuation has to be allowed. We show that the eavesdropper can make use of this property to obtain a remarkable part of the secret key by replacing some of the DI nonlocal boxes with local ones. On the contrary, the same cheating strategy does not apply to the device-dependent (DD) version of the protocol. Thus such kind of DI protocol is less secure than its DD counterpart., Comment: 4 pages

We further study the security of the quantum bit commitment (QBC) protocol we previously proposed [Phys. Rev. A 74, 022332 (2006).], by analyzing the reduced density matrix \rho_{b}^{B} which describes the quantum state at Bob's side corresponding to Alice's committed bit b. It is shown that Alice will find \rho_{0}^{B}\perp \rho_{1}^{B} while the protocol remains concealing to Bob. On the contrary, the existing no-go theorem of unconditionally secure QBC is based on the condition \rho_{0}^{B}\simeq \rho_{1}^{B}. Thus the specific cheating strategy proposed in the no-go theorem does not necessarily applies to our protocol., Comment: 10 pages, 1 table

In a recent letter (Phys. Lett. A 377 (2013) 1076, arXiv:0905.3801), the authors presented an impossibility proof of quantum bit commitment, which attempted to cover all possible protocols that involve both quantum and classical information. Here we show that there are many errors in the proof, thus it fails to exhaust all conceivable protocols., Comment: 3 pages

We simplified our previously proposed quantum bit commitment (QBC) protocol based on the Mach-Zehnder interferometer, by replacing symmetric beam splitters with asymmetric ones. It eliminates the need for random sending time of the photons; thus, the feasibility and efficiency are both improved. The protocol is immune to the cheating strategy in the Mayers-Lo-Chau no-go theorem of unconditionally secure QBC, because the density matrices of the committed states do not satisfy a crucial condition on which the no-go theorem holds., Comment: Published version

In a recent paper (Phys. Rev. Lett. 109, 160501 (2012). arXiv:1201.0849), it is claimed that any quantum protocol for classical two-sided computation between Alice and Bob can be proven completely insecure for Alice if it is secure against Bob. Here we show that the proof is not sufficiently general, because the security definition it based on is only a sufficient condition but not a necessary condition., Comment: 1.5 pages. Comment on the paper PRL 109, 160501 (2012), which also appeared as arXiv:1201.0849. The locations of the original content cited in this comment are referred to those in the PRL version

While unconditionally secure bit commitment (BC) is considered impossible within the quantum framework, it can be obtained under relativistic or experimental constraints. Here we study whether such BC can lead to secure quantum oblivious transfer (QOT). The answer is not completely negative. On one hand, we provide a detailed cheating strategy, showing that the "honest-but-curious adversaries" in some of the existing no-go proofs on QOT still apply even if secure BC is used, enabling the receiver to increase the average reliability of the decoded value of the transferred bit. On the other hand, it is also found that some other no-go proofs claiming that a dishonest receiver can always decode all transferred bits simultaneously with reliability 100% become invalid in this scenario, because their models of cryptographic protocols are too ideal to cover such a BC-based QOT., Comment: Published version. This paper generalized some results in Sec. V of arXiv:1101.4587, and pointed out the limitation of the proof in arXiv:quant-ph/9611031

Recently there were many quantum protocols devoted to solve the millionaire problem and private comparison problem by adding a semi-honest third party. They all require complicated quantum methods, while still leak a non-trivial amount of information to at least one of the parties. But it will be shown here that once the third party is introduced, there are very simple protocols which require quantum key distribution as the only quantum resource, and the amount of information leaked can be made arbitrarily small. Furthermore, even a dishonest third party cannot spoil the protocols. Thus our solutions surpass all existing protocols on both feasibility and security., Comment: The protocol is further improved so that it remains secure even if the third party is distrustful

For more than a decade, it was believed that unconditionally secure quantum bit commitment (QBC) is impossible. But basing on a previously proposed quantum key distribution scheme using orthogonal states, here we build a QBC protocol in which the density matrices of the quantum states encoding the commitment do not satisfy a crucial condition on which the no-go proofs of QBC are based. Thus the no-go proofs could be evaded. Our protocol is fault-tolerant and very feasible with currently available technology. It reopens the venue for other "post-cold-war" multi-party cryptographic protocols, e.g., quantum bit string commitment and quantum strong coin tossing with an arbitrarily small bias. This result also has a strong influence on the Clifton-Bub-Halvorson theorem which suggests that quantum theory could be characterized in terms of information-theoretic constraints., Comment: Published version plus an appendix showing how to defeat the counterfactual attack, more references [76,77,90,118-120] cited, and other minor changes

By proposing a paradox between the impossibility of superluminal signal transfer and the normalization condition of wavefunctions, we predict that when a change happens to the conditions that determining the status of a quantum system, the system will show no response to this change at all, until after a certain time interval. Otherwise either special relativity or quantum mechanics will be violated. As a consequence, no physical process can actually happen within Planck time. Therefore time is discrete, with Planck time being the smallest unit. More intriguingly, systems with a larger size and a slower speed will have a larger unit of time. Unlike many other interpretations of the discreteness of time, our proof can be tested, at less partly, by experiments. Our result also sets a limit on the speed of computers, and gives instruction to the search of quantum gravity theories., Comment: Added an appendix on another solution of the evolution of the wavefunction, which satisfies the normalization condition but still violates quantum mechanics

When the boundary condition of a quantum system changes, how fast will it affect the state of the system? Here we show that if the response takes place immediately, then it can allow superluminal signal transfer. Else if the response propagates in space with a finite speed, then it could give a simple explanation why our world shows classicality on the macroscopic scale. Furthermore, determining the exact value of this speed can either clarify the doubts on static experiments for testing Bell's inequality, or support the pilot-wave interpretation of quantum mechanics. We propose an experimental scheme for measuring this speed, which can be implemented with state-of-art technology, e.g., single-electron biprism interference., Comment: Title changed to suit the content more faithfully. The proposed experimental scheme remains unchanged. Added suggested specifications for implementing the scheme with single-electron biprism interference

It is revealed that quantum repeating devices and quantum seals have a very close relationship, thus the theory in one field can be applied to the other. Consequently, it is shown that the fidelity bounds and optimality of quantum repeating devices can be violated when they are used for decoding classical information from quantum states, and security bounds for protocols sealing quantum data exist., Comment: Published version

The relationship between the quantum bit commitment (QBC) and quantum seal (QS) is studied. It is elaborated that QBC and QS are not equivalent, but QS protocols satisfying a stronger unconditional security requirement can lead to an unconditionally secure QBC. In this sense, QS is strictly stronger than QBC in security requirements. Based on an earlier proposal on sealing a single bit, a feasible QBC protocol is also put forward, which is secure in practice even if the cheater has a strong quantum computational power., Comment: 6 pages, 1 figure

It was indicated [Yu 2007 Phys. Rev. A 75 066301] that a previous proposed quantum secret sharing (QSS) protocol based on Smolin states [Augusiak 2006 Phys. Rev. A 73 012318] is insecure against an internal cheater. Here we build a different QSS protocol with Smolin states alone, and prove it to be secure against known cheating strategies. Thus we open a promising venue for building secure QSS using merely Smolin states, which is a typical kind of bound entangled states. We also propose a feasible scheme to implement the protocol experimentally., Comment: Published version

Analyzing carefully an experimentally feasible non-entangled single qubit quantum secret sharing protocol and its modified version [Phys. Rev. Lett. 95, 230505 (2005); ibid. 98, 028902 (2007)], it is found that both versions are insecure against coherent attacks though the original idea is so remarkable. To overcome this fatal flaw, here we propose a protocol with a distinct security checking strategy, which still involves single qubit operations only, making it possible to achieve better security of quantum secret sharing with current technology., Comment: Published version

Recently an attack strategy was proposed by Chau [H. F. Chau, quant-ph/0602099 v3], which was claimed to be able to break all quantum string seal protocols, including the one proposed by He [G. P. He, Int. J. Quant. Inform. 4, 677 (2006)]. Here it will be shown that the information obtained in He's protocol by the attack is trivial. Thus Chau's conclusion that all quantum string seals are insecure is wrong. It will also be shown that some other claims in Chau's paper are inaccurate either., Comment: 2 pages

It was claimed that all quantum string seals are insecure [H. F. Chau, quant-ph/0602099]. However, here it will be shown that for imperfect quantum string seals, the information obtained by the measurement proposed in that reference is trivial. Therefore imperfect quantum string seals can be unconditionally secure., Comment: 2 pages

A scheme is proposed which stores classical data in 4-state quantum registers. It can achieve the following goal: the classical data can always be read unambiguously, while the quantum registers cannot be copied. Therefore the data provider can always distinguish the original quantum registers from piracy copies. Examples of application are also given., Comment: 5 pages

Though all-or-nothing oblivious transfer and one-out-of-two oblivious transfer are equivalent in classical cryptography, we here show that due to the nature of quantum cryptography, a protocol built upon secure quantum all-or-nothing oblivious transfer cannot satisfy the rigorous definition of quantum one-out-of-two oblivious transfer., Comment: 4 pages, no figure

The model of the quantum protocols sealing a classical bit is studied. It is shown that there exist upper bounds on its security. For any protocol where the bit can be read correctly with the probability $\alpha $, and reading the bit can be detected with the probability $\beta $, the upper bounds are $% \beta \leqslant 1/2$ and $\alpha +\beta \leqslant 9/8$., Comment: 5 pages, 1 figure. Add an example on how to break the majority voting scheme

Though it was proven that secure quantum sealing of a single classical bit is impossible in principle, here we propose an unconditionally secure quantum sealing protocol which seals a classical bit string. Any reader can obtain each bit of the sealed string with an arbitrarily small error rate, while reading the string is detectable. The protocol is simple and easy to be implemented. The possibility of using this protocol to seal a single bit in practical is also discussed., Comment: Add a discussion on the possibility of sealing a single bit in practical

Based on quantum entanglement, an all-or-nothing oblivious transfer protocol is proposed and is proven to be secure. The distinct merit of the present protocol lies in that it is not based on quantum bit commitment. More intriguingly, this OT protocol does not belong to a class of protocols denied by the Lo's no-go theorem of one-sided two-party secure computation, and thus its security can be achieved., Comment: 9 pages, 1 figure

An experimental scheme is proposed to test Bell's inequality by using superconducting nanocircuits. In this scheme, quantum entanglement of a pair of charge qubits separated in a sufficient long distance may be created by cavity quantum electrodynamic techniques; the population of qubits is experimentally measurable by dc currents through the probe junctions, and one measured outcome may be recorded for every experiment. Therefore, both locality and detection efficiency loopholes should be closed in the same experiment. We also propose a useful method to measure the amount of entanglement based on the concurrence between Josephson qubits. The measurable variables for Bell's inequality as well as the entanglement are expressed in terms of a useful phase-space Q function., Comment: 6 pages; Typos in Eqs. (8) and (9) in the published version were corrected

Based on the fact that the entanglement can not be created locally, we proposed a quantum bit commitment protocol, in which entangled states and quantum algorithms is used. The bit is not encoded with the form of the quantum states, and delaying the measurement is required. Therefore the protocol will not be denied by the Mayers-Lo-Chau no-go theorem, and unconditional security is achieved., Comment: This revised version added a brief discussion on how the protocol evades the MLC theorem

Using the recursive Green's function technique, we study the coherent electron conductance of a quantum point contact in the presence of a scanning probe microscope tip. Images of the coherent fringe inside a quantum point contact for different widths are obtained. It is found that the conductance of a specific channel is reduced while other channels are not affected as long as the tip is located at the positions correspending to that channel. Moreover, the coherent fringe is smoothed out by increasing the temperature or the voltage across the device. Our results are consistent with the experiments reported by Topinka et al.[Science 289, 2323 (2000)]., Comment: 5 pages