arXiv subjects

Ning Zhang

Publications and source records attributed to Ning Zhang.

At least 217 records · Page 12Linked to original sources

arXiv2022

PieTrack: An MOT solution based on synthetic data training and self-supervised domain adaptation

In order to cope with the increasing demand for labeling data and privacy issues with human detection, synthetic data has been used as a substitute and showing promising results in human detection and tracking tasks. We participate in the 7th Workshop on Benchmarking Multi-Target Tracking (BMTT), themed on "How Far Can Synthetic Data Take us"? Our solution, PieTrack, is developed based on synthetic data without using any pre-trained weights. We propose a self-supervised domain adaptation method that enables mitigating the domain shift issue between the synthetic (e.g., MOTSynth) and real data (e.g., MOT17) without involving extra human labels. By leveraging the proposed multi-scale ensemble inference, we achieved a final HOTA score of 58.7 on the MOT17 testing set, ranked third place in the challenge.

cs.CV↗

arXiv2022

Gridless Multisnapshot Variational Line Spectral Estimation from Coarsely Quantized Samples

Due to the increasing demand for low power and higher sampling rates, low resolution quantization for data acquisition has drawn great attention recently. Consequently, line spectral estimation (LSE) with multiple measurement vectors (MMVs) from coarsely quantized samples is of vital importance in cutting edge array signal processing applications such as range estimation and DOA estimation in millimeter wave radar systems. In this paper, we combine the low complexity gridless variational line spectral estimation (VALSE) and expectation propagation (EP) and propose an MVALSE-EP algorithm to estimate the frequencies from coarsely quantized samples. In addition, the Cramér Rao bound (CRB) is derived as a benchmark performance of the proposed algorithm, and insights are provided to reveal the effects of system parameters on estimation performance. It is shown that snapshots benefits the frequency estimation, especially in coarsely quantized scenarios. Numerical experiments are conducted to demonstrate the effectiveness of MVALSE-EP, including real data set.

eess.SP↗

arXiv2022

FadMan: Federated Anomaly Detection across Multiple Attributed Networks

Anomaly subgraph detection has been widely used in various applications, ranging from cyber attack in computer networks to malicious activities in social networks. Despite an increasing need for federated anomaly detection across multiple attributed networks, only a limited number of approaches are available for this problem. Federated anomaly detection faces two major challenges. One is that isolated data in most industries are restricted share with others for data privacy and security. The other is most of the centralized approaches training based on data integration. The main idea of federated anomaly detection is aligning private anomalies from local data owners on the public anomalies from the attributed network in the server through public anomalies to federate local anomalies. In each private attributed network, the detected anomaly subgraph is aligned with an anomaly subgraph in the public attributed network. The significant public anomaly subgraphs are selected for federated private anomalies while preventing local private data leakage. The proposed algorithm FadMan is a vertical federated learning framework for public node aligned with many private nodes of different features, and is validated on two tasks correlated anomaly detection on multiple attributed networks and anomaly detection on an attributeless network using five real-world datasets. In the first scenario, FadMan outperforms competitive methods by at least 12% accuracy at 10% noise level. In the second scenario, by analyzing the distribution of abnormal nodes, we find that the nodes of traffic anomalies are associated with the event of postgraduate entrance examination on the same day.

cs.LG↗

arXiv2022

Integration of Blockchain and Edge Computing in Internet of Things: A Survey

As an important technology to ensure data security, consistency, traceability, etc., blockchain has been increasingly used in Internet of Things (IoT) applications. The integration of blockchain and edge computing can further improve the resource utilization in terms of network, computing, storage, and security. This paper aims to present a survey on the integration of blockchain and edge computing. In particular, we first give an overview of blockchain and edge computing. We then present a general architecture of an integration of blockchain and edge computing system. We next study how to utilize blockchain to benefit edge computing, as well as how to use edge computing to benefit blockchain. We also discuss the issues brought by the integration of blockchain and edge computing system and solutions from perspectives of resource management, joint optimization, data management, computation offloading and security mechanism. Finally, we analyze and summarize the existing challenges posed by the integration of blockchain and edge computing system and the potential solutions in the future.

cs.CR↗

arXiv2022

Temperature uncertainty relation in non-equilibrium thermodynamics

Temperature uncertainty of a quantum system in canonical ensemble is inversely determined by its energy fluctuation, which is known as the temperature-energy uncertainty relation. No such uncertainty relation was discovered for a non-equilibrium open quantum system. In this article, we derive a universal temperature uncertainty relation for general non-equilibrium processes. We find that it is the fluctuation of heat, which is defined as the change in bath energy, determines the temperature uncertainty in non-equilibrium case. Specifically, the heat is divided into trajectory heat and backaction heat, which are associated with the system's trajectory of evolution and the backaction of partial measurement on system, respectively. Based on this decomposition, we reveal that both correlations between system and bath's process function and state function are the resources for enhancing temperature precision. Our findings are conductive to design ultrahigh sensitive quantum thermometer.

quant-ph↗

arXiv2022

Nuclear spin self compensation system for moving MEG sensing with optical pumped atomic spin co-magnetometer

Recording the moving MEGs of a person in which a person's head could move freely as we record the brain's magnetic field is a hot topic in recent years. Traditionally, atomic magnetometers are utilized for moving MEGs recording and a large compensation coil system is utilized for background magnetic field compensation. Here we described a new potential candidate: an optically pumped atomic co-magnetometer(OPACM) for moving MEGs recording. In the OPACM, hyper-polarized nuclear spins could produce a magnetic field which will shield the background fluctuation low frequency magnetic field noise while the the fast changing MEGs signal could be recorded. The nuclear spins look like an automatic magnetic field shields and dynamically compensate the fluctuated background magnetic field noise. In this article, the magnetic field compensation is studied theoretically and we find that the compensation is closely related to several parameters such as the electron spin magnetic field, the nuclear spin magnetic field and the holding magnetic field. Based on the model, the magnetic field compensation could be optimized. We also experimentally studied the magnetic field compensation and the responses of the OPACM to different frequencies of magnetic field are measured. We show that the OPACM owns a clear suppression of low frequency magnetic field under 1Hz and response to magnetic field's frequencies around the band of the MEGs. Magnetic field sensitivity of $3fT/Hz^{1/2}$ has been achieved. Finally, we do a simulation for the OPACM as it is utilized for moving MEGs recording. For comparison, the traditional compensation system for moving MEGs recording is based on a coil which is around 2m in dimension while our compensation system is only 2mm in dimension. Moreover, our compensation system could work in situ and will not affect each other.

physics.app-ph↗

arXiv2022

Stability of China's Stock Market: Measure and Forecast by Ricci Curvature on Network

The systemic stability of a stock market is one of the core issues in the financial field. The market can be regarded as a complex network whose nodes are stocks connected by edges that signify their correlation strength. Since the market is a strongly nonlinear system, it is difficult to measure the macroscopic stability and depict market fluctuations in time. In this paper, we use a geometric measure derived from discrete Ricci curvature to capture the higher-order nonlinear architecture of financial networks. In order to confirm the effectiveness of our method, we use it to analyze the CSI 300 constituents of China's stock market from 2005--2020 and the systemic stability of the market is quantified through the network's Ricci type curvatures. Furthermore, we use a hybrid model to analyze the curvature time series and predict the future trends of the market accurately. As far as we know, this is the first paper to apply Ricci curvature to forecast the systemic stability of domestic stock market, and our results show that Ricci curvature has good explanatory power for the market stability and can be a good indicator to judge the future risk and volatility of the domestic market.

q-fin.ST↗

arXiv2022

New Upper Bounds on the Error Probability under ML Decoding for Spinal Codes and the Joint Transmission-Decoding System Design

Spinal codes are a type of capacity-achieving rateless codes that have been proved to approach the Shannon capacity over the additive white Gaussian noise (AWGN) channel and the binary symmetric channel (BSC). In this paper, we aim to analyze the bounds on the error probability of Spinal codes and design a joint transmission-decoding system. First, in the finite block-length regime, we derive new upper bounds on the Maximum Likelihood (ML) decoding error probability for Spinal codes over both the AWGN channel and the BSC. Then, based on the derived bounds, we formulate a rate maximization problem. As the solution exhibits an incremental-tail-transmission pattern, we propose an improved transmission scheme, referred to as the thresholded incremental tail transmission (TITT) scheme. Moreover, we also develop a dynamic TITT-matching decoding algorithm, called the bubble decoding with memory (BD-M) algorithm, to reduce the decoding time complexity. The TITT scheme at the transmitter and the BD-M algorithm at the receiver jointly constitute a dynamic transmission-decoding system for Spinal code, improving its rate performance and decoding throughput. Theoretical analysis and simulation results are provided to verify the superiority of the derived bounds and the proposed joint transmission-decoding system design.

cs.IT↗

arXiv2022

Age of Information with Hybrid-ARQ: A Unified Explicit Result

Delivering timely status updates in a timeliness-critical communication system is of paramount importance to assist accurate and efficient decision making. Therefore, the topic of analyzing Age of Information has aroused new research interest. This paper contributes to new results in this area by systematically analyzing the AoI of two types of Hybrid Automatic Repeat reQuest (HARQ) techniques that have been newly standardized in the Release-16 5G New Radio (NR) specifications, namely reactive HARQ and proactive HARQ. Under a code-based status update system with non-trivial coding delay, transmission delay, propagation delay, decoding delay, and feedback delay, we derive unified closed-form average AoI and average Peak AoI expressions for reactive HARQ and proactive HARQ, respectively. Based on the obtained explicit expressions, we formulate an AoI minimization problem to investigate the age-optimal codeblock assignment strategy in the finite block-length (FBL) regime. Through case studies and analytical results, we provide comparative insights between reactive HARQ and proactive HARQ from a perspective of freshness of information. The numerical results and optimization solutions show that proactive HARQ draws its strength from both age performance and system robustness, thus enabling the potential to provide new system advancement of a freshness-critical status update system.

cs.IT↗

arXiv2022

Non-Markovian quantum thermometry

The rapidly developing quantum technologies and thermodynamics have put forward a requirement to precisely control and measure the temperature of microscopic matter at the quantum level. Many quantum thermometry schemes have been proposed. However, precisely measuring low temperature is still challenging because the obtained sensing errors generally tend to diverge with decreasing temperature. Using a continuous-variable system as a thermometer, we propose non-Markovian quantum thermometry to measure the temperature of a quantum reservoir. A mechanism to make the sensing error $δT$ scale with the temperature $T$ as the Landau bound $δT\simeq T$ in the full-temperature regime is discovered. Our analysis reveals that it is the quantum criticality of the total thermometer-reservoir system that causes this enhanced sensitivity. Efficiently avoiding the error-divergence problem, our result gives an efficient way to precisely measure the low temperature of quantum systems.

quant-ph↗

arXiv2022

Mobile Wireless Rechargeable UAV Networks: Challenges and Solutions

Unmanned aerial vehicles (UAVs) can help facilitate cost-effective and flexible service provisioning in future smart cities. Nevertheless, UAV applications generally suffer severe flight time limitations due to constrained onboard battery capacity, causing a necessity of frequent battery recharging or replacement when performing persistent missions. Utilizing wireless mobile chargers, such as vehicles with wireless charging equipment for on-demand self-recharging has been envisioned as a promising solution to address this issue. In this article, we present a comprehensive study of \underline{v}ehicle-assisted \underline{w}ireless rechargeable \underline{U}AV \underline{n}etworks (VWUNs) to promote on-demand, secure, and efficient UAV recharging services. Specifically, we first discuss the opportunities and challenges of deploying VWUNs and review state-of-the-art solutions in this field. We then propose a secure and privacy-preserving VWUN framework for UAVs and ground vehicles based on differential privacy (DP). Within this framework, an online double auction mechanism is developed for optimal charging scheduling, and a two-phase DP algorithm is devised to preserve the sensitive bidding and energy trading information of participants. Experimental results demonstrate that the proposed framework can effectively enhance charging efficiency and security. Finally, we outline promising directions for future research in this emerging field.

eess.SP↗

arXiv2022

Ulam floating functions

We extend the notion of Ulam floating sets from convex bodies to Ulam floating functions. We use the Ulam floating functions to derive a new variational formula for the affine surface area of log-concave functions.

math.MG↗

arXiv2022

A Multi-factor Multi-level and Interaction based (M2I) Authentication Framework for Internet of Things (IoT) Applications

Existing authentication solutions proposed for Internet of Things (IoT) provide a single Level of Assurance (LoA) regardless of the sensitivity levels of the resources or interactions between IoT devices being protected. For effective (with adequate level of protection) and efficient (with as low overhead costs as possible) protections, it may be desirable to tailor the protection level in response to the sensitivity level of the resources, as a stronger protection level typically imposes a higher level of overheads costs. In this paper, we investigate how to facilitate multi-LoA authentication for IoT by proposing a multi-factor multi-level and interaction based (M2I) authentication framework. The framework implements LoA linked and interaction based authentication. Two interaction modes are investigated, P2P (Peer-to-Peer) and O2M (One-to-Many) via the design of two corresponding protocols. Evaluation results show that adopting the O2M interaction mode in authentication can cut communication cost significantly; compared with that of the Kerberos protocol, the O2M protocol reduces the communication cost by 42% ~ 45%. The protocols also introduce less computational cost. The P2P and O2M protocol, respectively, reduce the computational cost by 70% ~ 72% and 81% ~ 82% in comparison with that of Kerberos. Evaluation results also show that the two factor authentication option costs twice as much as that of the one-factor option.

cs.CR↗

arXiv2022

Hierarchical BERT for Medical Document Understanding

Medical document understanding has gained much attention recently. One representative task is the International Classification of Disease (ICD) diagnosis code assignment. Existing work adopts either RNN or CNN as the backbone network because the vanilla BERT cannot handle well long documents (>2000 to kens). One issue shared across all these approaches is that they are over specific to the ICD code assignment task, losing generality to give the whole document-level and sentence-level embedding. As a result, it is not straight-forward to direct them to other downstream NLU tasks. Motivated by these observations, we propose Medical Document BERT (MDBERT) for long medical document understanding tasks. MDBERT is not only effective in learning representations at different levels of semantics but efficient in encoding long documents by leveraging a bottom-up hierarchical architecture. Compared to vanilla BERT solutions: 1, MDBERT boosts the performance up to relatively 20% on the MIMIC-III dataset, making it comparable to current SOTA solutions; 2, it cuts the computational complexity on self-attention modules to less than 1/100. Other than the ICD code assignment, we conduct a variety of other NLU tasks on a large commercial dataset named as TrialTrove, to showcase MDBERT's strength in delivering different levels of semantics.

cs.CL↗

arXiv2022

LoopITR: Combining Dual and Cross Encoder Architectures for Image-Text Retrieval

Dual encoders and cross encoders have been widely used for image-text retrieval. Between the two, the dual encoder encodes the image and text independently followed by a dot product, while the cross encoder jointly feeds image and text as the input and performs dense multi-modal fusion. These two architectures are typically modeled separately without interaction. In this work, we propose LoopITR, which combines them in the same network for joint learning. Specifically, we let the dual encoder provide hard negatives to the cross encoder, and use the more discriminative cross encoder to distill its predictions back to the dual encoder. Both steps are efficiently performed together in the same model. Our work centers on empirical analyses of this combined architecture, putting the main focus on the design of the distillation objective. Our experimental results highlight the benefits of training the two encoders in the same network, and demonstrate that distillation can be quite effective with just a few hard negative examples. Experiments on two standard datasets (Flickr30K and COCO) show our approach achieves state-of-the-art dual encoder performance when compared with approaches using a similar amount of data.

cs.CV↗

arXiv2022

The Picard group of the loop space of the Riemann sphere

The loop space of the Riemann sphere consisting of all $C^k$ or Sobolev $W^{k,p}$ maps from the circle $S^1$ to the sphere is an infinite dimensional complex manifold. We compute the Picard group of holomorphic line bundles on this loop space as an infinite dimensional complex Lie group with Lie algebra the first Dolbeault group. The group of Mobius transformations $G$ and its loop group $LG$ act on this loop space. We prove that an element of the Picard group is $LG$-fixed if it is $G$-fixed; thus completely answer the question by Millson and Zombro about $G$-equivariant projective embedding of the loop space of the Riemann sphere.

math.CV↗

arXiv2022

Holomorphic self-maps of the loop space of $\mathbb{P}^n$

The loop space $L\mathbb{P}^n$ of the complex projective space $\mathbb{P}^n$ consisting of all $C^k$ or Sobolev $W^{k, \, p}$ maps $S^1 \to \mathbb{P}^n$ is an infinite dimensional complex manifold. We identify a class of holomorphic self-maps of $L\mathbb{P}^n$, including all automorphisms.

math.CV↗

arXiv2022

Holomorphic sections of line bundles on the jet spaces of the Riemann sphere

Fix a point $t_0$ in the circle $S^1$. The space $J^k(t_0, \mathbb{P}^1)$ of $k$-jets at $t_0$ of $C^{\infty}$ maps from $S^1$ to the Riemann sphere $\mathbb{P}^1$ is a $k+1$ dimensional complex algebraic manifold. We identify a class of holomorphic sections of line bundles on $J^k(t_0, \mathbb{P}^1)$.

math.CV↗