arXiv subjects

Rui Zhang

Publications and source records attributed to Rui Zhang.

At least 991 records · Page 55Linked to original sources

arXiv2020

Delay-Optimal Scheduling for IRS-Aided Mobile Edge Computing

We consider an intelligent reflecting surface (IRS)-aided mobile edge computing (MEC) system, where an IRS is deployed to assist computation offloading from two users to an access point connected with an edge cloud. For the IRS-aided data transmission, in contrast to the conventional non-orthogonal multiple-access (NOMA) and time-division multiple-access (TDMA), we propose a new flexible time-sharing NOMA scheme that allows users to flexibly divide their data into two parts transmitted via NOMA and TDMA, respectively, thus encapsulating both conventional NOMA and TDMA as special cases. We formulate an optimization problem to minimize the sum delay of the two users by designing the IRS passive reflection and users' computation-offloading scheduling under the IRS discrete-phase constraint. Although this problem is non-convex, we obtain its optimal solution for both the cases of infinite and finite cloud computing capacities. Furthermore, we show that NOMA and TDMA based transmissions are preferred in different scenarios, depending on the users' cloud-computing time as well as the rate discrepancy between NOMA and TDMA with IRS.

cs.IT↗

arXiv2020

A Single-species Atomic Comagnetometer Based on 87Rb atoms

The comagnetometer has been one of the most sensitive devices with which to test new physics related to spin-dependent interactions, but the comagnetometers based on overlapping ensembles of multiple spin species usually suffer from systematic errors due to magnetic field gradients. Here, we propose a comagnetometer based on the Zeeman transitions of the dual hyperfine levels in ground-state 87Rb atoms, which shows nearly negligible sensitivity to variations of laser power and frequency, magnetic field, and magnetic field gradients. We measured the hypothetical spin-dependent gravitational energy of the proton with the comagnetometer, which is smaller than 4*10^{-18} eV, comparable to the most stringent existing constraint. Through optimization of the atomic cell, it is possible to improve the accuracy of the comagnetometer further.

physics.atom-ph↗

arXiv2020

Quantile Propagation for Wasserstein-Approximate Gaussian Processes

Approximate inference techniques are the cornerstone of probabilistic methods based on Gaussian process priors. Despite this, most work approximately optimizes standard divergence measures such as the Kullback-Leibler (KL) divergence, which lack the basic desiderata for the task at hand, while chiefly offering merely technical convenience. We develop a new approximate inference method for Gaussian process models which overcomes the technical challenges arising from abandoning these convenient divergences. Our method---dubbed Quantile Propagation (QP)---is similar to expectation propagation (EP) but minimizes the $L_2$ Wasserstein distance (WD) instead of the KL divergence. The WD exhibits all the required properties of a distance metric, while respecting the geometry of the underlying sample space. We show that QP matches quantile functions rather than moments as in EP and has the same mean update but a smaller variance update than EP, thereby alleviating EP's tendency to over-estimate posterior variances. Crucially, despite the significant complexity of dealing with the WD, QP has the same favorable locality property as EP, and thereby admits an efficient algorithm. Experiments on classification and Poisson regression show that QP outperforms both EP and variational Bayes.

cs.LG↗

arXiv2020

The extremal process of super-Brownian motion

In this paper, we establish limit theorems for the supremum of the support, denoted by $M_t$, of a supercritical super-Brownian motion $\{X_t, t\ge0\}$ on $\mathbb{R}$. We prove that there exists an $m(t)$ such that $(X_t-m(t), M_t-m(t))$ converges in law, and give some large deviation results for $M_t$ as $t\to\infty$. We also prove that the limit of the extremal process $\mathcal{E}_t:=X_t-m(t)$ is a Poisson random measure with exponential intensity in which each atom is decorated by an independent copy of an auxiliary measure. These results are analogues of the results for branching Brownian motions obtained in Arguin et al. (Probab. Theory Relat. Fields 157 (2013), 535-574), Aïdékon et al. (Probab. Theory Relat. Fields 157 (2013), 405-451) and Roberts (Ann. Probab. 41 (2013), 3518-3541).

math.PR↗

arXiv2020

3D Placement for Multi-UAV Relaying: An Iterative Gibbs-Sampling and Block Coordinate Descent Optimization Approach

In this paper, we consider an unmanned aerial vehicle (UAV) enabled relaying system where multiple UAVs are deployed as aerial relays to support simultaneous communications from a set of source nodes to their destination nodes on the ground. An optimization problem is formulated under practical channel models to maximize the minimum achievable expected rate among all pairs of ground nodes by jointly designing UAVs' three-dimensional (3D) placement as well as the bandwidth-and-power allocation. This problem, however, is non-convex and thus difficult to solve. As such, we propose a new method, called iterative Gibbs-sampling and block-coordinate-descent (IGS-BCD), to efficiently obtain a high-quality suboptimal solution by synergizing the advantages of both the deterministic (BCD) and stochastic (GS) optimization methods. Specifically, our proposed method alternates between two optimization phases until convergence is reached, namely, one phase that uses the BCD method to find locally-optimal UAVs' 3D placement and the other phase that leverages the GS method to generate new UAVs' 3D placement for exploration. Moreover, we present an efficient method for properly initializing UAVs' placement that leads to faster convergence of the proposed IGS-BCD algorithm. Numerical results show that the proposed IGS-BCD and initialization methods outperform the conventional BCD or GS method alone in terms of convergence-and-performance trade-off, as well as other benchmark schemes.

cs.IT↗

arXiv2020

Cooperative Beam Routing for Multi-IRS Aided Communication

Intelligent reflecting surface (IRS) has been deemed as a transformative technology to achieve smart and reconfigurable environment for wireless communication. This letter studies a new IRS-aided communication system, where multiple IRSs assist in the communication between a multi-antenna base station (BS) and a remote single-antenna user by multi-hop signal reflection. Specifically, by exploiting the line-of-sight (LoS) link between nearby IRSs, a multi-hop cascaded LoS link between the BS and user is established where a set of IRSs are selected to successively reflect the BS's signal, so that the received signal power at the user is maximized. To tackle this new problem, we first present the closed-form solutions for the optimal active and cooperative passive beamforming at the BS and selected IRSs, respectively, for a given beam route. Then, we derive the end-to-end channel power, which unveils a fundamental trade-off in the optimal beam routing design between maximizing the multiplicative passive beamforming gain and minimizing the multi-reflection path loss. To reconcile this trade-off, we recast the IRS selection and beam routing problem as an equivalent shortest simple-path problem in graph theory and solve it optimally. Numerical results show significant performance gains of the proposed algorithm over benchmark schemes and also draw useful insights into the optimal beam routing design.

cs.IT↗

arXiv2020

Did You Ask a Good Question? A Cross-Domain Question Intention Classification Benchmark for Text-to-SQL

Neural models have achieved significant results on the text-to-SQL task, in which most current work assumes all the input questions are legal and generates a SQL query for any input. However, in the real scenario, users can input any text that may not be able to be answered by a SQL query. In this work, we propose TriageSQL, the first cross-domain text-to-SQL question intention classification benchmark that requires models to distinguish four types of unanswerable questions from answerable questions. The baseline RoBERTa model achieves a 60% F1 score on the test set, demonstrating the need for further improvement on this task. Our dataset is available at https://github.com/chatc/TriageSQL.

cs.CL↗

arXiv2020

Isovector parton distribution functions of the proton on a superfine lattice

We study isovector unpolarized and helicity parton distribution functions (PDF) of the proton within the framework of Large Momentum Effective Theory. We use a gauge ensemble, generated by the MILC Collaboration, with a superfine lattice spacing of $0.042$ fm and a pion mass of $310$ MeV, enabling us to simultaneously reach sub-fermi spatial separations and larger nucleon momenta. We compare the spatial dependence of quasi-PDF matrix elements in different renormalization schemes with the corresponding results of the global fits, obtained using 1-loop perturbative matching. We present determinations of the first four moments of the unpolarized and helicity PDFs of proton from the Ioffe-time dependence of the isovector matrix elements, obtained by employing a ratio-based renormalization scheme.

hep-lat↗

arXiv2020

UAV Trajectory and Communication Co-design: Flexible Path Discretization and Path Compression

The performance optimization of UAV communication systems requires the joint design of UAV trajectory and communication efficiently. To tackle the challenge of infinite design variables arising from the continuous-time UAV trajectory optimization, a commonly adopted approach is by approximating the UAV trajectory with piecewise-linear path segments in three-dimensional (3D) space. However, this approach may still incur prohibitive computational complexity in practice when the UAV flight period/distance becomes long, as the distance between consecutive waypoints needs to be kept sufficiently small to retain high approximation accuracy. To resolve this fundamental issue, we propose in this paper a new and general framework for UAV trajectory and communication co-design. First, we propose a flexible path discretization scheme that optimizes only a number of selected waypoints (designable waypoints) along the UAV path for complexity reduction, while all the designable and non-designable waypoints are used in calculating the approximated communication utility along the UAV trajectory for ensuring high trajectory discretization accuracy. Next, given any number of designable waypoints, we propose a novel path compression scheme where the UAV 3D path is first decomposed into three one-dimensional (1D) sub-paths and each sub-path is then approximated by superimposing a number of selected basis paths weighted by their corresponding path coefficients, thus further reducing the path design complexity. Finally, we provide a case study on UAV trajectory design for aerial data harvesting from distributed ground sensors, and numerically show that the proposed schemes can significantly reduce the UAV trajectory design complexity yet achieve favorable rate performance as compared to conventional path/time discretization schemes.

cs.IT↗

arXiv2020

Uplink Channel Estimation for Double-IRS Assisted Multi-User MIMO

To achieve the more promising passive beamforming gains in the double-intelligent reflecting surface (IRS) assisted system over the conventional single-IRS system, channel estimation is practically indispensable but also a more challenging problem to tackle, due to the presence of not only the single- but also double-reflection links that are intricately coupled. In this paper, we propose a new and efficient channel estimation scheme for the double-IRS assisted uplink multiple-input multiple-output (MIMO) communication system to resolve the cascaded channel state information (CSI) of both its single- and double-reflection links. First, for the single-user case, the higher-dimensional double-reflection channel is efficiently estimated at the multi-antenna base station (BS) with low training overhead by exploiting the fact that its cascaded channel coefficients are scaled versions of those of a lower-dimensional single-reflection channel. Then, the proposed channel estimation scheme is extended to the multi-user case, where given an arbitrary user's cascaded channel estimated as in the single-user case, the other users' cascaded channels are scaled versions of it and thus can be estimated with reduced training overhead. Simulation results verify the effectiveness of the proposed channel estimation scheme as compared to the benchmark scheme.

cs.IT↗

arXiv2020

Massive MIMO for Cellular-Connected UAV: Challenges and Promising Solutions

Massive multiple-input multiple-output (MIMO) is a promising technology for enabling cellular-connected unmanned aerial vehicle (UAV) communications in the future. Equipped with full-dimensional large arrays, ground base stations (GBSs) can apply adaptive fine-grained three-dimensional (3D) beamforming to mitigate the strong interference between high-altitude UAVs and low-altitude terrestrial users, thus significantly enhancing the network spectral efficiency. However, the performance gain of massive MIMO critically depends on the accurate channel state information (CSI) of both UAVs and terrestrial users at the GBSs, which is practically difficult to achieve due to UAV-induced pilot contamination and UAV's high mobility in 3D. Moreover, the increasingly popular applications relying on a large group of coordinated UAVs or UAV swarm as well as the practical hybrid GBS beamforming architecture for massive MIMO further complicate the pilot contamination and channel/beam tracking problems. In this article, we provide an overview of the above challenging issues, propose new solutions to cope with them, and discuss about promising directions for future research. Preliminary simulation results are also provided to validate the effectiveness of proposed solutions.

cs.IT↗

arXiv2020

MALA: Cross-Domain Dialogue Generation with Action Learning

Response generation for task-oriented dialogues involves two basic components: dialogue planning and surface realization. These two components, however, have a discrepancy in their objectives, i.e., task completion and language quality. To deal with such discrepancy, conditioned response generation has been introduced where the generation process is factorized into action decision and language generation via explicit action representations. To obtain action representations, recent studies learn latent actions in an unsupervised manner based on the utterance lexical similarity. Such an action learning approach is prone to diversities of language surfaces, which may impinge task completion and language quality. To address this issue, we propose multi-stage adaptive latent action learning (MALA) that learns semantic latent actions by distinguishing the effects of utterances on dialogue progress. We model the utterance effect using the transition of dialogue states caused by the utterance and develop a semantic similarity measurement that estimates whether utterances have similar effects. For learning semantic actions on domains without dialogue states, MsALA extends the semantic similarity measurement across domains progressively, i.e., from aligning shared actions to learning domain-specific actions. Experiments using multi-domain datasets, SMD and MultiWOZ, show that our proposed model achieves consistent improvements over the baselines models in terms of both task completion and language quality.

cs.CL↗

arXiv2020

Generalizable and Explainable Dialogue Generation via Explicit Action Learning

Response generation for task-oriented dialogues implicitly optimizes two objectives at the same time: task completion and language quality. Conditioned response generation serves as an effective approach to separately and better optimize these two objectives. Such an approach relies on system action annotations which are expensive to obtain. To alleviate the need of action annotations, latent action learning is introduced to map each utterance to a latent representation. However, this approach is prone to over-dependence on the training data, and the generalization capability is thus restricted. To address this issue, we propose to learn natural language actions that represent utterances as a span of words. This explicit action representation promotes generalization via the compositional structure of language. It also enables an explainable generation process. Our proposed unsupervised approach learns a memory component to summarize system utterances into a short span of words. To further promote a compact action representation, we propose an auxiliary task that restores state annotations as the summarized dialogue context using the memory component. Our proposed approach outperforms latent action baselines on MultiWOZ, a benchmark multi-domain dataset.

cs.CL↗

arXiv2020

GraphDialog: Integrating Graph Knowledge into End-to-End Task-Oriented Dialogue Systems

End-to-end task-oriented dialogue systems aim to generate system responses directly from plain text inputs. There are two challenges for such systems: one is how to effectively incorporate external knowledge bases (KBs) into the learning framework; the other is how to accurately capture the semantics of dialogue history. In this paper, we address these two challenges by exploiting the graph structural information in the knowledge base and in the dependency parsing tree of the dialogue. To effectively leverage the structural information in dialogue history, we propose a new recurrent cell architecture which allows representation learning on graphs. To exploit the relations between entities in KBs, the model combines multi-hop reasoning ability based on the graph structure. Experimental results show that the proposed model achieves consistent improvement over state-of-the-art models on two different task-oriented dialogue datasets.

cs.CL↗

arXiv2020

Rank one tensor completion problem

In this paper, we consider the rank-one tensor completion problem. We address the question of existence and uniqueness of the rank-one solution. In particular we show that the global uniqueness over the field of real numbers can be verified in a polynomial time. We give examples showing that there is an essential difference between the question of global uniqueness over the fields of real and complex numbers. Finally we briefly discuss the rank-one approximation problem for noisy observations.

math.NA↗

arXiv2020

Intelligent Reflecting Surface Assisted Multi-User OFDMA: Channel Estimation and Training Design

To achieve the full passive beamforming gains of intelligent reflecting surface (IRS), accurate channel state information (CSI) is indispensable but practically challenging to acquire, due to the excessive amount of channel parameters to be estimated which increases with the number of IRS reflecting elements as well as that of IRS-served users. To tackle this challenge, we propose in this paper two efficient channel estimation schemes for different channel setups in an IRS-assisted multi-user broadband communication system employing the orthogonal frequency division multiple access (OFDMA). The first channel estimation scheme, which estimates the CSI of all users in parallel simultaneously at the access point (AP), is applicable for arbitrary frequency-selective fading channels. In contrast, the second channel estimation scheme, which exploits a key property that all users share the same (common) IRS-AP channel to enhance the training efficiency and support more users, is proposed for the typical scenario with line-of-sight (LoS) dominant user-IRS channels. For the two proposed channel estimation schemes, we further optimize their corresponding training designs (including pilot tone allocations for all users and IRS time-varying reflection pattern) to minimize the channel estimation error. Moreover, we derive and compare the fundamental limits on the minimum training overhead and the maximum number of supportable users of these two schemes. Simulation results verify the effectiveness of the proposed channel estimation schemes and training designs, and show their significant performance improvement over various benchmark schemes.

cs.IT↗

arXiv2020

Generative Image Inpainting with Submanifold Alignment

Image inpainting aims at restoring missing regions of corrupted images, which has many applications such as image restoration and object removal. However, current GAN-based generative inpainting models do not explicitly exploit the structural or textural consistency between restored contents and their surrounding contexts.To address this limitation, we propose to enforce the alignment (or closeness) between the local data submanifolds (or subspaces) around restored images and those around the original (uncorrupted) images during the learning process of GAN-based inpainting models. We exploit Local Intrinsic Dimensionality (LID) to measure, in deep feature space, the alignment between data submanifolds learned by a GAN model and those of the original data, from a perspective of both images (denoted as iLID) and local patches (denoted as pLID) of images. We then apply iLID and pLID as regularizations for GAN-based inpainting models to encourage two levels of submanifold alignment: 1) an image-level alignment for improving structural consistency, and 2) a patch-level alignment for improving textural details. Experimental results on four benchmark datasets show that our proposed model can generate more accurate results than state-of-the-art models.

cs.CV↗

arXiv2020

Short-Term and Long-Term Context Aggregation Network for Video Inpainting

Video inpainting aims to restore missing regions of a video and has many applications such as video editing and object removal. However, existing methods either suffer from inaccurate short-term context aggregation or rarely explore long-term frame information. In this work, we present a novel context aggregation network to effectively exploit both short-term and long-term frame information for video inpainting. In the encoding stage, we propose boundary-aware short-term context aggregation, which aligns and aggregates, from neighbor frames, local regions that are closely related to the boundary context of missing regions into the target frame. Furthermore, we propose dynamic long-term context aggregation to globally refine the feature map generated in the encoding stage using long-term frame features, which are dynamically updated throughout the inpainting process. Experiments show that it outperforms state-of-the-art methods with better inpainting results and fast inpainting speed.

cs.CV↗