arXiv subjects

Yu Cai

Publications and source records attributed to Yu Cai.

54 records · Page 3Linked to original sources

arXiv2018

Geometry of the set of quantum correlations

It is well known that correlations predicted by quantum mechanics cannot be explained by any classical (local-realistic) theory. The relative strength of quantum and classical correlations is usually studied in the context of Bell inequalities, but this tells us little about the geometry of the quantum set of correlations. In other words, we do not have good intuition about what the quantum set actually looks like. In this paper we study the geometry of the quantum set using standard tools from convex geometry. We find explicit examples of rather counter-intuitive features in the simplest non-trivial Bell scenario (two parties, two inputs and two outputs) and illustrate them using 2-dimensional slice plots. We also show that even more complex features appear in Bell scenarios with more inputs or more parties. Finally, we discuss the limitations that the geometry of the quantum set imposes on the task of self-testing.

quant-ph↗

arXiv2018

Errors in Flash-Memory-Based Solid-State Drives: Analysis, Mitigation, and Recovery

NAND flash memory is ubiquitous in everyday life today because its capacity has continuously increased and cost has continuously decreased over decades. This positive growth is a result of two key trends: (1) effective process technology scaling; and (2) multi-level (e.g., MLC, TLC) cell data coding. Unfortunately, the reliability of raw data stored in flash memory has also continued to become more difficult to ensure, because these two trends lead to (1) fewer electrons in the flash memory cell floating gate to represent the data; and (2) larger cell-to-cell interference and disturbance effects. Without mitigation, worsening reliability can reduce the lifetime of NAND flash memory. As a result, flash memory controllers in solid-state drives (SSDs) have become much more sophisticated: they incorporate many effective techniques to ensure the correct interpretation of noisy data stored in flash memory cells. In this chapter, we review recent advances in SSD error characterization, mitigation, and data recovery techniques for reliability and lifetime improvement. We provide rigorous experimental data from state-of-the-art MLC and TLC NAND flash devices on various types of flash memory errors, to motivate the need for such techniques. Based on the understanding developed by the experimental characterization, we describe several mitigation and recovery techniques, including (1) cell-tocell interference mitigation; (2) optimal multi-level cell sensing; (3) error correction using state-of-the-art algorithms and methods; and (4) data recovery when error correction fails. We quantify the reliability improvement provided by each of these techniques. Looking forward, we briefly discuss how flash memory and these techniques could evolve into the future.

cs.AR↗

arXiv2017

Promise of Commercialization: Carbon Materials for Low-Cost Perovskite Solar Cells

Perovskite solar cells (PVSCs) have attracted extensive studies due to their high power conversion efficiency (PCE) with low-cost in both raw material and processes. However, there remain obstacles that hinder the way to its commercialization. Among many drawbacks in PVSCs, we note the problems brought by the use of noble metal counter electrodes (CEs) such as gold (Au) and silver (Ag). The costly Au and Ag need high energy-consumption thermal evaporation process which can be made only with expensive evaporation equipment under vacuum. All the factors elevate the threshold of PVSCs' commercialization. Carbon material, on the other hand, is a readily available electrode candidate for the application as CE in the PVSCs. In this review, endeavors on PVSCs with low-cost carbon materials will be comprehensively discussed based on different device structures and carbon composition. We believe that the PVSCs with carbon-based CE hold the promise of commercialization of this new technology.

physics.app-ph↗

arXiv2017

Many-box locality

There is an ongoing search for a physical or operational definition for quantum mechanics. Several informational principles have been proposed which are satisfied by a theory less restrictive than quantum mechanics. Here, we introduce the principle of "many-box locality", which is a refined version of the previously proposed "macroscopic locality". These principles are based on coarse-graining the statistics of several copies of a given box. The set of behaviors satisfying many-box locality for $N$ boxes is denoted $MBL_N$. We study these sets in the bipartite scenario with two binary measurements, in relation with the sets $\mathcal{Q}$ and $\mathcal{Q}_{1+AB}$ of quantum and "almost quantum" correlations. We find that the $MBL_N$ sets are in general not convex. For unbiased marginals, by working in the Fourier space we can prove analytically that $MBL_{N}\subsetneq\mathcal{Q}$ for any finite $N$, while $MBL_{\infty}=\mathcal{Q}$. Then, with suitably developed numerical tools, we find an example of a point that belongs to $MBL_{16}$ but not to $\mathcal{Q}_{1+AB}$. Among the problems that remain open, is whether $\mathcal{Q}\subset MBL_{\infty}$.

quant-ph↗

arXiv2017

Error Characterization, Mitigation, and Recovery in Flash Memory Based Solid-State Drives

NAND flash memory is ubiquitous in everyday life today because its capacity has continuously increased and cost has continuously decreased over decades. This positive growth is a result of two key trends: (1) effective process technology scaling, and (2) multi-level (e.g., MLC, TLC) cell data coding. Unfortunately, the reliability of raw data stored in flash memory has also continued to become more difficult to ensure, because these two trends lead to (1) fewer electrons in the flash memory cell (floating gate) to represent the data and (2) larger cell-to-cell interference and disturbance effects. Without mitigation, worsening reliability can reduce the lifetime of NAND flash memory. As a result, flash memory controllers in solid-state drives (SSDs) have become much more sophisticated: they incorporate many effective techniques to ensure the correct interpretation of noisy data stored in flash memory cells. In this article, we review recent advances in SSD error characterization, mitigation, and data recovery techniques for reliability and lifetime improvement. We provide rigorous experimental data from state-of-the-art MLC and TLC NAND flash devices on various types of flash memory errors, to motivate the need for such techniques. Based on the understanding developed by the experimental characterization, we describe several mitigation and recovery techniques, including (1) cell-to-cell interference mitigation, (2) optimal multi-level cell sensing, (3) error correction using state-of-the-art algorithms and methods, and (4) data recovery when error correction fails. We quantify the reliability improvement provided by each of these techniques. Looking forward, we briefly discuss how flash memory and these techniques could evolve into the future.

cs.AR↗

arXiv2017

Witnessing irreducible dimension

The Hilbert space dimension of a quantum system is the most basic quantifier of its information content. Lower bounds on the dimension can be certified in a device-independent way, based only on observed statistics. We highlight that some such "dimension witnesses" capture only the presence of systems of some dimension, which in a sense is trivial, not the capacity of performing information processing on them, which is the point of experimental efforts to control high-dimensional systems. In order to capture this aspect, we introduce the notion of irreducible dimension of a quantum behaviour. This dimension can be certified, and we provide a witness for irreducible dimension four.

quant-ph↗

arXiv2017

Experimental many-pairs nonlocality

Collective measurements on large quantum systems together with a majority voting strategy can lead to a violation of the CHSH Bell inequality. In presence of many entangled pairs, this violation decreases quickly with the number of pairs, and vanishes for some critical pair number that is a function of the noise present in the system. Here, we show that a different binning strategy can lead to a more substantial Bell violation when the noise is sufficiently small. Given the relation between the critical pair number and the source noise, we then present an experiment where the critical pair number is used to quantify the quality of a high visibility photon pair source. Our results demonstrate nonlocal correlations using collective measurements operating on clusters of more than 40 photon pairs.

quant-ph↗

arXiv2016

A new device-independent dimension witness and its experimental implementation

A dimension witness is a criterion that sets a lower bound on the dimension needed to reproduce the observed data. Three types of dimension witnesses can be found in the literature: device-dependent ones, in which the bound is obtained assuming some knowledge on the state and the measurements; device-independent prepare-and-measure ones, that can be applied to any system including classical ones; and device-independent Bell-based ones, that certify the minimal dimension of some entangled systems. Here we consider the Collins-Gisin-Linden-Massar-Popescu (CGLMP) Bell-type inequality for four outcomes. We show that a sufficiently high violation of this inequality witnesses $d \geq 4$ and present a proof-of-principle experimental observation of such a violation. This presents a first experimental violation of the third type of dimension witness beyond qutrits.

quant-ph↗

arXiv2016

Measurement-dependent locality beyond i.i.d

When conducting a Bell test, it is normal to assume that the preparation of the quantum state is independent of the measurements performed on it. Remarkably, the violation of local realism by entangled quantum systems can be certified even if this assumption is partially relaxed. Here, we allow such measurement dependence to correlate multiple runs of the experiment, going beyond previous studies that considered independent and identically distributed (i.i.d.) runs. To do so, we study the polytope that defines block-i.i.d. measurement-dependent local models. We prove that non-i.i.d. models are strictly more powerful than i.i.d. ones, and comment on the relevance of this work for the study of randomness amplification in simple Bell scenarios with suitably optimised inequalities.

quant-ph↗

arXiv2016

Sparse Coding and Counting for Robust Visual Tracking

In this paper, we propose a novel sparse coding and counting method under Bayesian framwork for visual tracking. In contrast to existing methods, the proposed method employs the combination of L0 and L1 norm to regularize the linear coefficients of incrementally updated linear basis. The sparsity constraint enables the tracker to effectively handle difficult challenges, such as occlusion or image corruption. To achieve realtime processing, we propose a fast and efficient numerical algorithm for solving the proposed model. Although it is an NP-hard problem, the proposed accelerated proximal gradient (APG) approach is guaranteed to converge to a solution quickly. Besides, we provide a closed solution of combining L0 and L1 regularized representation to obtain better sparsity. Experimental results on challenging video sequences demonstrate that the proposed method achieves state-of-the-art results both in accuracy and speed.

cs.CV↗

arXiv2015

State complexity and quantum computation

The complexity of a quantum state may be closely related to the usefulness of the state for quantum computation. We discuss this link using the tree size of a multiqubit state, a complexity measure that has two noticeable (and, so far, unique) features: it is in principle computable, and non-trivial lower bounds can be obtained, hence identifying truly complex states. In this paper, we first review the definition of tree size, together with known results on the most complex three and four qubits states. Moving to the multiqubit case, we revisit a mathematical theorem for proving a lower bound on tree size that scales superpolynomially in the number of qubits. Next, states with superpolynomial tree size, the Immanant states, the Deutsch-Jozsa states, the Shor's states and the subgroup states, are described. We show that the universal resource state for measurement based quantum computation, the 2D-cluster state, has superpolynomial tree size. Moreover, we show how the complexity of subgroup states and the 2D cluster state can be verified efficiently. The question of how tree size is related to the speed up achieved in quantum computation is also addressed. We show that superpolynomial tree size of the resource state is essential for measurement based quantum computation. The necessary role of large tree size in the circuit model of quantum computation is still a conjecture; and we prove a weaker version of the conjecture.

quant-ph↗

arXiv2015

Aqueous Proton Transfer Across Single Layer Graphene

Proton transfer across single layer graphene is associated with large computed energy barriers and is therefore thought to be unfavorable at room temperature unless nanoscale holes or dopants are introduced, or a potential bias is applied. Here, we subject single layer graphene supported on fused silica to cycles of high and low pH and show that protons transfer reversibly from the aqueous phase through the graphene to the other side where they undergo acid-base chemistry with the silica hydroxyl groups. After ruling out diffusion through macroscopic pinholes, the protons are found to transfer through rare, naturally occurring atomic defects. Computer simulations reveal low energy barriers of 0.68 to 0.75 eV for aqueous proton transfer across hydroxyl-terminated atomic defects that participate in a Grotthuss-type relay, while pyrylium-like ether terminations shut down proton exchange. Unfavorable energy barriers to helium and hydrogen transfer indicate the transfer process is selective for aqueous protons.

cond-mat.mtrl-sci↗

arXiv2014

Robust self testing of the 3-qubit $W$ state

Self-testing is a device independent method which can be used to determine the nature of a physical system or device, without knowing any detail of the inner mechanism or the physical dimension of Hilbert space of the system. The only information required are the number of measurements, number of outputs of each measurement and the statistics of each measurement. Earlier works on self testing restricted either to two parties scenario or multipartite graph states. Here, we construct a method to self-test the three-qubit $W$ state, and show how to extend it to other pure three-qubit states. Our bounds are robust against the inevitable experimental errors.

quant-ph↗

arXiv2014

Maximal tree size of few-qubit states

Tree size ($\rm{TS}$) is an interesting measure of complexity for multiqubit states: not only is it in principle computable, but one can obtain lower bounds for it. In this way, it has been possible to identify families of states whose complexity scales superpolynomially in the number of qubits. With the goal of progressing in the systematic study of the mathematical property of $\rm{TS}$, in this work we characterize the tree size of pure states for the case where the number of qubits is small, namely, 3 or 4. The study of three qubits does not hold great surprises, insofar as the structure of entanglement is rather simple; the maximal $\rm{TS}$ is found to be 8, reached for instance by the $|\rm{W}\rangle$ state. The study of four qubits yields several insights: in particular, the most economic description of a state is found not to be recursive. The maximal $\rm{TS}$ is found to be 16, reached for instance by a state called $|Ψ^{(4)}\rangle$ which was already discussed in the context of four-photon down-conversion experiments. We also find that the states with maximal tree size form a set of zero measure: a smoothed version of tree size over a neighborhood of a state ($ε-\rm{TS}$) reduces the maximal values to 6 and 14, respectively. Finally, we introduce a notion of tree size for mixed states and discuss it for a one-parameter family of states.

quant-ph↗

arXiv2013

Tree-size complexity of multiqubit states

Complexity is often invoked alongside size and mass as a characteristic of macroscopic quantum objects. In 2004, Aaronson introduced the \textit{tree size} (TS) as a computable measure of complexity and studied its basic properties. In this paper, we improve and expand on those initial results. In particular, we give explicit characterizations of a family of states with superpolynomial complexity $n^{Ω(\log n)}= \mathrm{TS} =O(\sqrt{n}!)$ in the number of qubits $n$; and we show that any matrix-product state whose tensors are of dimension $D\times D$ has polynomial complexity $\mathrm{TS}=O(n^{\log_2 2D})$.

quant-ph↗

arXiv1995

Intrinsic Regularization Method in QCD

There exist certain intrinsic relations between the ultraviolet divergent graphs and the convergent ones at the same loop order in renormalizable quantum field theories. Whereupon we may establish a new method, the intrinsic regularization method, to regularize those divergent graphs. In this paper, we apply this method to QCD at the one loop order. It turns out to be satisfactory:The gauge invariance is preserved manifestly and the results are the same as those derived by means of other regularization methods.

hep-th↗

arXiv1994

On inserter regularization method

There exist certain intrinsic relations between the ultraviolet divergent graphs and the convergent ones at the same loop order in renormalizable quantum field theories. Whereupon we present a new method, the inserter regularization method, to regulate those divergent graphs. In this letter, we demonstrate this method with the $ϕ^4$ theory and QED at the one loop order. Some applications to SUSY-models are also made at the one loop order, which shows that supersymmetry is preserved manifestly and consistently.

hep-th↗

arXiv1994

A Note On Intrinsic Regularization Method

There exist certain intrinsic relations between the ultraviolet divergent graphs and the convergent ones at the same loop order in renormalizable quantum field theories. Whereupon we may establish a new method, the intrinsic regularization method, to regulate those divergent graphs. In this note, we present a proposal, the inserter proposal, to the method. The $ϕ^4$ theory and QED at the one loop order are dealt with in some detail. Inserters in the standard model are given. Some applications to SUSY-models are also made at the one loop order.

hep-th↗