arXiv subjects

Jun Zhang

Publications and source records attributed to Jun Zhang.

At least 829 records · Page 46Linked to original sources

arXiv2021

Pyramidal Dense Attention Networks for Lightweight Image Super-Resolution

Recently, deep convolutional neural network methods have achieved an excellent performance in image superresolution (SR), but they can not be easily applied to embedded devices due to large memory cost. To solve this problem, we propose a pyramidal dense attention network (PDAN) for lightweight image super-resolution in this paper. In our method, the proposed pyramidal dense learning can gradually increase the width of the densely connected layer inside a pyramidal dense block to extract deep features efficiently. Meanwhile, the adaptive group convolution that the number of groups grows linearly with dense convolutional layers is introduced to relieve the parameter explosion. Besides, we also present a novel joint attention to capture cross-dimension interaction between the spatial dimensions and channel dimension in an efficient way for providing rich discriminative feature representations. Extensive experimental results show that our method achieves superior performance in comparison with the state-of-the-art lightweight SR methods.

eess.IV↗

arXiv2021

HMM-Free Encoder Pre-Training for Streaming RNN Transducer

This work describes an encoder pre-training procedure using frame-wise label to improve the training of streaming recurrent neural network transducer (RNN-T) model. Streaming RNN-T trained from scratch usually performs worse than non-streaming RNN-T. Although it is common to address this issue through pre-training components of RNN-T with other criteria or frame-wise alignment guidance, the alignment is not easily available in end-to-end manner. In this work, frame-wise alignment, used to pre-train streaming RNN-T's encoder, is generated without using a HMM-based system. Therefore an all-neural framework equipping HMM-free encoder pre-training is constructed. This is achieved by expanding the spikes of CTC model to their left/right blank frames, and two expanding strategies are proposed. To our best knowledge, this is the first work to simulate HMM-based frame-wise label using CTC model for pre-training. Experiments conducted on LibriSpeech and MLS English tasks show the proposed pre-training procedure, compared with random initialization, reduces the WER by relatively 5%~11% and the emission latency by 60 ms. Besides, the method is lexicon-free, so it is friendly to new languages without manually designed lexicon.

eess.AS↗

arXiv2021

AI Empowered Resource Management for Future Wireless Networks

Resource management plays a pivotal role in wireless networks, which, unfortunately, leads to challenging NP-hard problems. Artificial Intelligence (AI), especially deep learning techniques, has recently emerged as a disruptive technology to solve such challenging problems in a real-time manner. However, although promising results have been reported, practical design guidelines and performance guarantees of AI-based approaches are still missing. In this paper, we endeavor to address two fundamental questions: 1) What are the main advantages of AI-based methods compared with classical techniques; and 2) Which neural network should we choose for a given resource management task. For the first question, four advantages are identified and discussed. For the second question, \emph{optimality gap}, i.e., the gap to the optimal performance, is proposed as a measure for selecting model architectures, as well as, for enabling a theoretical comparison between different AI-based approaches. Specifically, for $K$-user interference management problem, we theoretically show that graph neural networks (GNNs) are superior to multi-layer perceptrons (MLPs), and the performance gap between these two methods grows with $\sqrt{K}$.

cs.IT↗

arXiv2021

Agile wide-field imaging with selective high resolution

Wide-field and high-resolution (HR) imaging is essential for various applications such as aviation reconnaissance, topographic mapping and safety monitoring. The existing techniques require a large-scale detector array to capture HR images of the whole field, resulting in high complexity and heavy cost. In this work, we report an agile wide-field imaging framework with selective high resolution that requires only two detectors. It builds on the statistical sparsity prior of natural scenes that the important targets locate only at small regions of interests (ROI), instead of the whole field. Under this assumption, we use a short-focal camera to image wide field with a certain low resolution, and use a long-focal camera to acquire the HR images of ROI. To automatically locate ROI in the wide field in real time, we propose an efficient deep-learning based multiscale registration method that is robust and blind to the large setting differences (focal, white balance, etc) between the two cameras. Using the registered location, the long-focal camera mounted on a gimbal enables real-time tracking of the ROI for continuous HR imaging. We demonstrated the novel imaging framework by building a proof-of-concept setup with only 1181 gram weight, and assembled it on an unmanned aerial vehicle for air-to-ground monitoring. Experiments show that the setup maintains 120$^{\circ}$ wide field-of-view (FOV) with selective 0.45$mrad$ instantaneous FOV.

cs.CV↗

arXiv2021

Information flow based defensive chain for data leakage detection and prevention: a survey

Mobile and IoT applications have greatly enriched our daily life by providing convenient and intelligent services. However, these smart applications have been a prime target of adversaries for stealing sensitive data. It poses a crucial threat to users' identity security, financial security, or even life security. Research communities and industries have proposed many Information Flow Control (IFC) techniques for data leakage detection and prevention, including secure modeling, type system, static analysis, dynamic analysis, \textit{etc}. According to the application's development life cycle, although most attacks are conducted during the application's execution phase, data leakage vulnerabilities have been introduced since the design phase. With a focus on lifecycle protection, this survey reviews the recent representative works adopted in different phases. We propose an information flow based defensive chain, which provides a new framework to systematically understand various IFC techniques for data leakage detection and prevention in Mobile and IoT applications. In line with the phases of the application life cycle, each reviewed work is comprehensively studied in terms of technique, performance, and limitation. Research challenges and future directions are also pointed out by consideration of the integrity of the defensive chain.

cs.CR↗

arXiv2021

Cores in discrete exchange economies with complex endowments

The core is a traditional and useful solution concept in economic theory. But in discrete exchange economies without transfers, when endowments are complex, the core may be empty. This motivates Balbuzanov and Kotowski (2019) to interpret endowments as exclusion rights and propose a new concept called exclusion core. Our contribution is twofold. First, we propose a rectification of the core to solve its problem under complex endowments. Second, we propose a refinement of Balbuzanov and Kotowski's exclusion core to improve its performance. Our two core concepts share a common idea of correcting the misused altruism of unaffected agents in blocking coalitions. We propose a mechanism to find allocations in the two cores.

econ.TH↗

arXiv2021

18.8 Gbps real-time quantum random number generator with a photonic integrated chip

Quantum random number generators (QRNGs) can produce true random numbers. Yet, the two most important QRNG parameters highly desired for practical applications, i.e., speed and size, have to be compromised during implementations. Here, we present the fastest and miniaturized QRNG with a record real-time output rate as high as 18.8 Gbps by combining a photonic integrated chip and the technology of optimized randomness extraction. We assemble the photonic integrated circuit designed for vacuum state QRNG implementation, InGaAs homodyne detector and high-bandwidth transimpedance amplifier into a single chip using hybrid packaging, which exhibits the excellent characteristics of integration and high-frequency response. With a sample rate of 2.5 GSa/s in a 10-bit analog-to-digital converter and subsequent paralleled postprocessing in a field programmable gate array, the QRNG outputs ultrafast random bitstreams via a fiber optic transceiver, whose real-time speed is validated in a personal computer.

quant-ph↗

arXiv2021

Privacy Inference Attacks and Defenses in Cloud-based Deep Neural Network: A Survey

Deep Neural Network (DNN), one of the most powerful machine learning algorithms, is increasingly leveraged to overcome the bottleneck of effectively exploring and analyzing massive data to boost advanced scientific development. It is not a surprise that cloud computing providers offer the cloud-based DNN as an out-of-the-box service. Though there are some benefits from the cloud-based DNN, the interaction mechanism among two or multiple entities in the cloud inevitably induces new privacy risks. This survey presents the most recent findings of privacy attacks and defenses appeared in cloud-based neural network services. We systematically and thoroughly review privacy attacks and defenses in the pipeline of cloud-based DNN service, i.e., data manipulation, training, and prediction. In particular, a new theory, called cloud-based ML privacy game, is extracted from the recently published literature to provide a deep understanding of state-of-the-art research. Finally, the challenges and future work are presented to help researchers to continue to push forward the competitions between privacy attackers and defenders.

cs.CR↗

arXiv2021

Hamiltonian knottedness and lifting paths from the shape invariant

The Hamiltonian shape invariant of a domain $X \subset \mathbb R^4$, as a subset of $\mathbb R^2$, describes the product Lagrangian tori which may be embedded in $X$. We provide necessary and sufficient conditions to determine whether or not a path in the shape invariant can lift, that is, be realized as a smooth family of embedded Lagrangian tori, when $X$ is a basic $4$-dimensional toric domain such as a ball $B^4(R)$, an ellipsoid $E(a,b)$ with $\frac{b}{a} \in {\mathbb N}_{\geq 2}$, or a polydisk $P(c,d)$. As applications, via the path lifting, we can detect knotted embeddings of product Lagrangian tori in many toric $X$. We also obtain novel obstructions to symplectic embeddings between domains that are more general than toric concave or toric convex.

math.SG↗

arXiv2021

Measurement-Device-Independent Verification of a Quantum Memory

In this paper we report an experiment that verifies an atomic-ensemble quantum memory via a measurement-device-independent scheme. A single photon generated via Rydberg blockade in one atomic ensemble is stored in another atomic ensemble via electromagnetically induced transparency. After storage for a long duration, this photon is retrieved and interfered with a second photon to perform joint Bell-state measurement (BSM). Quantum state for each photon is chosen based on a quantum random number generator respectively in each run. By evaluating correlations between the random states and BSM results, we certify that our memory is genuinely entanglement-preserving.

quant-ph↗

arXiv2021

Communication-Efficient Federated Learning with Dual-Side Low-Rank Compression

Federated learning (FL) is a promising and powerful approach for training deep learning models without sharing the raw data of clients. During the training process of FL, the central server and distributed clients need to exchange a vast amount of model information periodically. To address the challenge of communication-intensive training, we propose a new training method, referred to as federated learning with dual-side low-rank compression (FedDLR), where the deep learning model is compressed via low-rank approximations at both the server and client sides. The proposed FedDLR not only reduces the communication overhead during the training stage but also directly generates a compact model to speed up the inference process. We shall provide convergence analysis, investigate the influence of the key parameters, and empirically show that FedDLR outperforms the state-of-the-art solutions in terms of both the communication and computation efficiency.

cs.LG↗

arXiv2021

Triangulation and persistence: Algebra 101

This paper lays the foundations of triangulated persistence categories (TPC), which brings together persistence modules with the theory of triangulated categories. As a result we introduce several measurements and metrics on the set of objects of some triangulated categories. We also provide examples of TPCs coming from algebra, algebraic topology, microlocal sheaf theory and symplectic topology.

math.AT↗

arXiv2021

Affine-modeled video extraction from a single motion blurred image

A motion-blurred image is the temporal average of multiple sharp frames over the exposure time. Recovering these sharp video frames from a single blurred image is nontrivial, due to not only its strong ill-posedness, but also various types of complex motion in reality such as rotation and motion in depth. In this work, we report a generalized video extraction method using the affine motion modeling, enabling to tackle multiple types of complex motion and their mixing. In its workflow, the moving objects are first segemented in the alpha channel. This allows separate recovery of different objects with different motion. Then, we reduce the variable space by modeling each video clip as a series of affine transformations of a reference frame, and introduce the $l0$-norm total variation regularization to attenuate the ringing artifact. The differentiable affine operators are employed to realize gradient-descent optimization of the affine model, which follows a novel coarse-to-fine strategy to further reduce artifacts. As a result, both the affine parameters and sharp reference image are retrieved. They are finally input into stepwise affine transformation to recover the sharp video frames. The stepwise retrieval maintains the nature to bypass the frame order ambiguity. Experiments on both public datasets and real captured data validate the state-of-the-art performance of the reported technique.

cs.CV↗

arXiv2021

Large-scale phase retrieval

High-throughput computational imaging requires efficient processing algorithms to retrieve multi-dimensional and multi-scale information. In computational phase imaging, phase retrieval (PR) is required to reconstruct both amplitude and phase in complex space from intensity-only measurements. The existing PR algorithms suffer from the tradeoff among low computational complexity, robustness to measurement noise and strong generalization on different modalities. In this work, we report an efficient large-scale phase retrieval technique termed as LPR. It extends the plug-and-play generalized-alternating-projection framework from real space to nonlinear complex space. The alternating projection solver and enhancing neural network are respectively derived to tackle the measurement formation and statistical prior regularization. This framework compensates the shortcomings of each operator, so as to realize high-fidelity phase retrieval with low computational complexity and strong generalization. We applied the technique for a series of computational phase imaging modalities including coherent diffraction imaging, coded diffraction pattern imaging, and Fourier ptychographic microscopy. Extensive simulations and experiments validate that the technique outperforms the existing PR algorithms with as much as 17dB enhancement on signal-to-noise ratio, and more than one order-of-magnitude increased running efficiency. Besides, we for the first time demonstrate ultra-large-scale phase retrieval at the 8K level (7680$\times$4320 pixels) in minute-level time.

eess.IV↗

arXiv2021

Decentralized Statistical Inference with Unrolled Graph Neural Networks

In this paper, we investigate the decentralized statistical inference problem, where a network of agents cooperatively recover a (structured) vector from private noisy samples without centralized coordination. Existing optimization-based algorithms suffer from issues of model mismatch and poor convergence speed, and thus their performance would be degraded, provided that the number of communication rounds is limited. This motivates us to propose a learning-based framework, which unrolls well-noted decentralized optimization algorithms (e.g., Prox-DGD and PG-EXTRA) into graph neural networks (GNNs). By minimizing the recovery error via end-to-end training, this learning-based framework resolves the model mismatch issue. Our convergence analysis (with PG-EXTRA as the base algorithm) reveals that the learned model parameters may accelerate the convergence and reduce the recovery error to a large extent. The simulation results demonstrate that the proposed GNN-based learning methods prominently outperform several state-of-the-art optimization-based algorithms in convergence speed and recovery error.

cs.LG↗

arXiv2021

A Deep Neural Network Surrogate Modeling Benchmark for Temperature Field Prediction of Heat Source Layout

Thermal issue is of great importance during layout design of heat source components in systems engineering, especially for high functional-density products. Thermal analysis generally needs complex simulation, which leads to an unaffordable computational burden to layout optimization as it iteratively evaluates different schemes. Surrogate modeling is an effective way to alleviate computation complexity. However, temperature field prediction (TFP) with complex heat source layout (HSL) input is an ultra-high dimensional nonlinear regression problem, which brings great difficulty to traditional regression models. The Deep neural network (DNN) regression method is a feasible way for its good approximation performance. However, it faces great challenges in both data preparation for sample diversity and uniformity in the layout space with physical constraints, and proper DNN model selection and training for good generality, which necessitates efforts of both layout designer and DNN experts. To advance this cross-domain research, this paper proposes a DNN based HSL-TFP surrogate modeling task benchmark. With consideration for engineering applicability, sample generation, dataset evaluation, DNN model, and surrogate performance metrics, are thoroughly studied. Experiments are conducted with ten representative state-of-the-art DNN models. Detailed discussion on baseline results is provided and future prospects are analyzed for DNN based HSL-TFP tasks.

cs.LG↗

arXiv2021

Large System Achievable Rate Analysis of RIS-Assisted MIMO Wireless Communication with Statistical CSIT

Reconfigurable intelligent surface (RIS) is an emerging technology to enhance wireless communication in terms of energy cost and system performance by equipping a considerable quantity of nearly passive reflecting elements. This study focuses on a downlink RIS-assisted multiple-input multiple-output (MIMO) wireless communication system that comprises three communication links of Rician channel, including base station (BS) to RIS, RIS to user, and BS to user. The objective is to design an optimal transmit covariance matrix at BS and diagonal phase-shifting matrix at RIS to maximize the achievable ergodic rate by exploiting the statistical channel state information at BS. Therefore, a large-system approximation of the achievable ergodic rate is derived using the replica method in large dimension random matrix theory. This large-system approximation enables the identification of asymptotic-optimal transmit covariance and diagonal phase-shifting matrices using an alternating optimization algorithm. Simulation results show that the large-system results are consistent with the achievable ergodic rate calculated by Monte Carlo averaging. The results verify that the proposed algorithm can significantly enhance the RIS-assisted MIMO system performance.

cs.IT↗

arXiv2021

Reconstructing 3D Magnetic Topology of On-disk Prominence Bubbles from Stereoscopic Observations

Bubbles, the semi-circular voids below quiescent prominences (filaments), have been extensively investigated in the past decade. However, hitherto the magnetic nature of bubbles cannot be verified due to the lack of on-disk photospheric magnetic field observations. Here for the first time, we find and investigate an on-disk prominence bubble around a filament barb on 2019 March 18 based on stereoscopic observations from NVST, SDO, and STEREO-A. In high-resolution NVST H$α$ images, this bubble has a sharp arch-like boundary and a projected width of $\thicksim$26 Mm. Combining SDO and STEREO-A images, we further reconstruct 3D structure of the bubble boundary, whose maximum height is $\thicksim$15.6 Mm. The squashing factor Q map deduced from extrapolated 3D magnetic fields around the bubble depicts a distinct arch-shaped interface with a height of $\thicksim$11 Mm, which agrees well with the reconstructed 3D structure of the observed bubble boundary. Under the interface lies a set of magnetic loops, which is rooted on a surrounding photospheric magnetic patch. To be more persuasive, another on-disk bubble on 2019 June 10 is presented as a supplement. According to these results obtained from on-disk bubble observations, we suggest that the bubble boundary corresponds to the interface between the prominence dips (barb) and the underlying magnetic loops rooted nearby. It is thus reasonable to speculate that the bubble can form around a filament barb below which there is a photospheric magnetic patch.

astro-ph.SR↗