arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Dong Lin

415 papers · 1993–2026

1993: 11994: 11995: 11996: 01997: 31998: 11999: 22000: 32001: 52002: 162003: 72004: 62005: 112006: 102007: 132008: 132009: 132010: 212011: 252012: 172013: 202014: 132015: 152016: 162017: 152018: 162019: 132020: 262021: 172022: 192023: 192024: 162025: 202026: 20
1993peak 26 in 20202026

Frequent co-authors

Papers

showing 1–25 of 415 RSS

DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems

Ruoxi Wang, Rakesh Shivanna, Derek Z. Cheng +4 more

cs.IR · · cited 953× · arXiv ↗ · more like this ·

Learning effective feature crosses is the key behind building recommender systems. However, the sparse and large feature space requires exhaustive search to identify effective crosses. Deep & Cross Network (DCN) was proposed to automatically and efficiently learn bounded-degree predictive feature interactions. Unfortunately, in models that serve web-scale traffic with billions of training examples, DCN showed limited expressiveness in its cross network at learning more predictive feature interactions. Despite significant research progress made, many deep learning models in production still rely on traditional feed-forward neural networks to learn feature crosses inefficiently. In light of the pros/cons of DCN and existing feature interaction learning approaches, we propose an improved framework DCN-V2 to make DCN more practical in large-scale industrial settings. In a comprehensive…

Dynamics of random early detection

Dong Lin, Robert Morris

Computer Networks and Communications · · cited 794× · DOI ↗ · more like this ·

In this paper we evaluate the effectiveness of Random Early Detection (RED) over traffic types categorized as non-adaptive, fragile and robust, according to their responses to congestion. We point out that RED allows unfair bandwidth sharing when a mixture of the three traffic types shares a link. This unfairness is caused by the fact that at any given time RED imposes the same loss rate on all flows, regardless of their bandwidths.We propose Fair Random Early Drop (FRED), a modified version of RED. FRED uses per-active-flow accounting to impose on each flow a loss rate that depends on the flow's buffer use.We show that FRED provides better protection than RED for adaptive (fragile and robust) flows. In addition, FRED is able to isolate non-adaptive greedy traffic more effectively. Finally, we present a "two-packet-buffer" gateway mechanism to support a large number of flows without…

On the nonstationarity of Internet traffic

Jin Cao, William S. Cleveland, Dong Lin +1 more

Computer Networks and Communications · · cited 198× · DOI ↗ · more like this ·

Traffic variables on an uncongested Internet wire exhibit a pervasive nonstationarity. As the rate of new TCP connections increases, arrival processes (packet and connection) tend locally toward Poisson, and time series variables (packet sizes, transferred file sizes, and connection round-trip times) tend locally toward independent. The cause of the nonstationarity is superposition: the intermingling of sequences of connections between different source-destination pairs, and the intermingling of sequences of packets from different connections. We show this empirically by extensive study of packet traces for nine links coming from four packet header databases. We show it theoretically by invoking the mathematical theory of point processes and time series. If the connection rate on a link gets sufficiently high, the variables can be quite close to Poisson and independent; if major…

Understanding and Improving Knowledge Distillation

Jiaxi Tang, Rakesh Shivanna, Zhe Zhao +4 more

cs.LG · · cited 169× · arXiv ↗ · more like this ·

Knowledge Distillation (KD) is a model-agnostic technique to improve model quality while having a fixed capacity budget. It is a commonly used technique for model compression, where a larger capacity teacher model with better quality is used to train a more compact student model with better inference efficiency. Through distillation, one hopes to benefit from student's compactness, without sacrificing too much on model quality. Despite the large success of knowledge distillation, better understanding of how it benefits student model's training dynamics remains under-explored. In this paper, we categorize teacher's knowledge into three hierarchical levels and study its effects on knowledge distillation: (1) knowledge of the `universe', where KD brings a regularization effect through label smoothing; (2) domain knowledge, where teacher injects class relationships prior to student's logit…

A survey of error-concealment schemes for real-time audio and video transmissions over the Internet

Benjamin W. Wah, Xiao Su, Dong Lin

Signal Processing · · cited 155× · DOI ↗ · more like this ·

Real-time audio and video data streamed over unreliable IP networks, such as the Internet, may encounter losses due to dropped packets or late arrivals. This paper reviews error-concealment schemes developed for streaming real-time audio and video data over the Internet. Based on their interactions with (video or audio) source coders, we classify existing techniques into source coder-independent schemes that treat underlying source coders as black boxes, and source coder-dependent schemes that exploit coder-specific characteristics to perform reconstruction. Last, we identify possible future research directions.

Variance of aggregated Web traffic

Robert Morris, Dong Lin

Computer Networks and Communications · · cited 96× · DOI ↗ · more like this ·

If data traffic were Poisson, increases in the amount of traffic aggregated on a network would rapidly decrease the relative size of bursts. The discovery of pervasive long-range dependence demonstrates that real network traffic is burstier than any possible Poisson model. We present evidence that, despite being non-Poisson, aggregating Web traffic causes it to smooth out as rapidly as Poisson traffic. That is, the relationship between changes in mean bandwidth and changes in variance is the same for Web traffic as it is for Poisson traffic. We derive our evidence from traces of real traffic in two ways: first, by observing how variance changes over the large range of mean bandwidths present in 24-hour traces; second, by observing the relationship of variance and mean bandwidth for individual users and combinations of users. Our conclusion, that variance changes linearly with mean…

Travel behaviour and mobility needs of older adults in an ageing and car-dependent society

Jianqiang Cui, Becky P.Y. Loo, Dong Lin

Transportation · International Journal of Urban Sciences · · cited 89× · DOI ↗ · more like this ·

An ageing population poses long-term challenges for a nation’s health, labour force market, housing and transport, affecting economic growth, living standards and government finances. With increasingly rapid ageing in both developed and developing countries, how to develop future transport frameworks, taking into account the increasingly large ageing population and their travel needs, has been emerging as a major challenge in urban policy development. This review explores how demographic changes in an ageing society and the associated travel patterns of a changing older cohort may impact on future transport demand by providing evidence on the travel behaviour of older adults, its associated influencing factors and the possible alternatives to car transport for older adults. It concludes by discussing major policy areas for addressing the elderly’s mobility needs via the provision of…

A Comparative Study of Machine Learning Techniques for Credit Card Fraud Detection Based on Time Variance

Shantanu Rajora, Dong-Lin Li, Chandan Kumar Jha +5 more

Artificial Intelligence · UTS ePRESS (University of Technology Sydney) · · cited 74× · DOI ↗ · more like this ·

This paper proposes a comparative performance of ten different machine learning algorithms, done on a credit card fraud detection application. The machine learning methods have been classified into two groups namely classification algorithms and ensemble learning group. Each group is comprised of five different algorithms. Besides, the 'Time' feature is introduced in the data set and performances of the algorithms are studied with and without the 'Time' feature. Two algorithms of the ensemble learning group have been found to perform better when the used dataset does not include the 'Time' feature. However, for the classification algorithms group, three classifiers are found to show better predictive accuracies when all attributes are included in the used dataset. The rest of the machine learning models have approximate similar scores between these datasets.

Time delay signature concealment of optical feedback induced chaos in an external cavity semiconductor laser

Jiagui Wu, Guang-Qiong Xia, Xi Tang +4 more

Computer Networks and Communications · Optics Express · · cited 68× · DOI ↗ · more like this ·

The time delay (TD) signature concealment of optical feedback induced chaos in an external cavity semiconductor laser is experimentally demonstrated. Both the evolution curve and the distribution map of TD signature are obtained in the parameter space of external feedback strength and injection current. The optimum parameter scope of the TD signature concealment is also specified. Furthermore, the approximately periodic evolution relation between TD signature and external cavity length is observed and indicates that the intrinsic relaxation oscillation of semiconductor laser may play an important role during the process of TD signature suppression.

Robust Feature-Based Automated Multi-View Human Action Recognition System

Kuang-Pen Chou, Mukesh Prasad, Di Wu +6 more

Computer Vision and Pattern Recognition · IEEE Access · · cited 62× · DOI ↗ · more like this ·

Automated human action recognition has the potential to play an important role in public security, for example, in relation to the multiview surveillance videos taken in public places, such as train stations or airports. This paper compares three practical, reliable, and generic systems for multiview video-based human action recognition, namely, the nearest neighbor classifier, Gaussian mixture model classifier, and the nearest mean classifier. To describe the different actions performed in different views, view-invariant features are proposed to address multiview action recognition. These features are obtained by extracting the holistic features from different temporal scales which are modeled as points of interest which represent the global spatial-temporal distribution. Experiments and cross-data testing are conducted on the KTH, WEIZMANN, and MuHAVi datasets. The system does not…

Learning Multi-granular Quantized Embeddings for Large-Vocab Categorical Features in Recommender Systems

Wang-Cheng Kang, Derek Zhiyuan Cheng, Ting Chen +4 more

cs.IR · · cited 60× · arXiv ↗ · more like this ·

Recommender system models often represent various sparse features like users, items, and categorical features via embeddings. A standard approach is to map each unique feature value to an embedding vector. The size of the produced embedding table grows linearly with the size of the vocabulary. Therefore, a large vocabulary inevitably leads to a gigantic embedding table, creating two severe problems: (i) making model serving intractable in resource-constrained environments; (ii) causing overfitting problems. In this paper, we seek to learn highly compact embeddings for large-vocab sparse features in recommender systems (recsys). First, we show that the novel Differentiable Product Quantization (DPQ) approach can generalize to recsys problems. In addition, to better handle the power-law data distribution commonly seen in recsys, we propose a Multi-Granular Quantized Embeddings (MGQE)…

Quantitative Network Monitoring with NetQRE

Yifei Yuan, Dong Lin, Ankit Mishra +3 more

Computer Networks and Communications · · cited 56× · DOI ↗ · more like this ·

In network management today, dynamic updates are required for traffic engineering and for timely response to security threats. Decisions for such updates are based on monitoring network traffic to compute numerical quantities based on a variety of network and application-level performance metrics. Today's state-of-the-art tools lack programming abstractions that capture application or session-layer semantics, and thus require network operators to specify and reason about complex state machines and interactions across layers. To address this limitation, we present the design and implementation of NetQRE, a high-level declarative toolkit that aims to simplify the specification and implementation of such quantitative network policies. NetQRE integrates regular-expression-like pattern matching at flow-level as well as application-level payloads with aggregation operations such as sum and…

Optimal Contrast Grayscale Visual Cryptography Schemes With Reversing

Daoshun Wang, Tao Song, Dong Lin +1 more

Computer Vision and Pattern Recognition · IEEE Transactions on Information Forensics and Security · · cited 55× · DOI ↗ · more like this ·

The visual cryptography scheme (VCS) is an encryption technique that utilizes the human visual system in recovering a secret image and it does not require any complex calculation. However, the contrast of the reconstructed image could be quite low. A number of reversing-based VCSs (or VCSs with reversing) (RVCS) have been proposed for binary secret images, allowing participants to perform a reversing operation on shares (or shadows). This reversing operation can be easily implemented by current copy machines. Some existing traditional VCS schemes without reversing (nRVCS) can be extended to RVCS with the same pixel expansion for binary image, and the RVCS can achieve ideal contrast, significantly higher than that of the corresponding nRVCS. In the application of grayscale VCS, the contrast is much lower than that of the binary cases. Therefore, it is more desirable to improve the…

IP packet generation

William S. Cleveland, Dong Lin, Don X. Sun

Computer Networks and Communications · · cited 48× · DOI ↗ · more like this ·

TCP start times for HTTP are nonstationary. The nonstationarity occurs because the start times on a link, a point process, are a superposition of source traffic point processes, and the statistics of superposition changes as the number of superposed processes changes. The start time rate is a measure of the number of traffic sources. The univariate distribution of the inter-arrival times is approximately Weibull, and as the rate increases, the Weibull shape parameter goes to 1, an exponential distribution. The autocorrelation of the log inter-arrival times is described by a simple, two-parameter process: white noise plus a long-range persistent time series. As the rate increases, the variance of the persistent series tends to zero, so the log times tend to white noise. A parsimonious statistical model for log inter-arrivals accounts for the autocorrelation, the Weibull distribution, and…

Stable Propagation of Inhibited Spiking Dynamics in Vertical-Cavity Surface-Emitting Lasers for Neuromorphic Photonic Networks

Tao Deng, Joshua Robertson, Zheng-Mao Wu +5 more

Artificial Intelligence · IEEE Access · · cited 45× · DOI ↗ · more like this ·

We investigate experimentally and theoretically the communication of inhibited spiking dynamics between two interlinked photonic neurons based upon the vertical-cavity surface-emitting lasers (VCSELs). We show that the sub-nanosecond speed spiking dynamics fired by a Transmitter-VCSEL (T-VCSEL) can be inhibited under the arrival of suitable external stimuli and that the inhibited spiking behaviors are propagated into another Receiver-VCSEL (R-VCSEL). Both VCSELs exhibit analogous inhibited spiking dynamics in response to stimuli with different temporal durations and strength. In addition, a very good agreement is found between theoretical simulations and experiments. These results offer greater prospects for future networks of VCSEL-based photonic neurons for neuromorphic computing platforms.

A New Mechanism for Data Visualization with Tsk-Type Preprocessed Collaborative Fuzzy Rule Based System

Mukesh Prasad, Yuting Liu, Dong-Lin Li +3 more

Information Systems · Journal of Artificial Intelligence and Soft Computing Research · · cited 44× · DOI ↗ · more like this ·

Abstract A novel data knowledge representation with the combination of structure learning ability of preprocessed collaborative fuzzy clustering and fuzzy expert knowledge of Takagi- Sugeno-Kang type model is presented in this paper. The proposed method divides a huge dataset into two or more subsets of dataset. The subsets of dataset interact with each other through a collaborative mechanism in order to find some similar properties within each-other. The proposed method is useful in dealing with big data issues since it divides a huge dataset into subsets of dataset and finds common features among the subsets. The salient feature of the proposed method is that it uses a small subset of dataset and some common features instead of using the entire dataset and all the features. Before interactions among subsets of the dataset, the proposed method applies a mapping technique for granules…

Liquid: Unifying Nearline and Offline Big Data Integration

Raul Castro Fernandez, Peter Pietzuch, Jay Kreps +6 more

Information Systems · Conference on Innovative Data Systems Research · · cited 44× · OpenAlex ↗ · more like this ·

With more sophisticated data-parallel processing systems, the new bottleneck in data-intensive companies shifts from the back-end data systems to the data integration stack, which is responsible for the pre-processing of data for back-end applications. The use of back-end data systems with different access latencies and data integration requirements poses new challenges that current data integration stacks based on distributed file systems—proposed a decade ago for batch-oriented processing—cannot address. In this paper, we describe Liquid, a data integration stack that provides low latency data access to support near real-time in addition to batch applications. It supports incremental processing, and is cost-efficient and highly available. Liquid has two layers: a processing layer based on a stateful stream processing model, and a messaging layer with a highly-available…

Recent Developments on the Moment Problem

Gwo Dong Lin

math.PR · · cited 42× · arXiv ↗ · more like this ·

We consider univariate distributions with finite moments of all positive orders. The moment problem is to determine whether or not a given distribution is uniquely determined by the sequence of its moments. There is a huge literature on this classical topic. In this survey, we will focus only on the recent developments on the checkable moment-(in)determinacy criteria including Cramer's condition, Carleman's condition, Hardy's condition, Krein's condition and the growth rate of moments, which help us solve the problem more easily. Both Hamburger and Stieltjes cases are investigated. The former is concerned with distributions on the whole real line, while the latter deals only with distributions on the right half-line. Some new results or new simple (direct) proofs of previous criteria are provided. Finally, we review the most recent moment problem for products of independent random…

Evaluating Image Tracking Approaches for Surface Velocimetry With Thermal Tracers

Dong Lin, Jens Grundmann, Anette Eltner

Computer Vision and Pattern Recognition · Water Resources Research · · cited 39× · DOI ↗ · more like this ·

Abstract In this paper an automatic approach is proposed to measure flow velocity with an uncooled thermal camera. Hot water is used as thermal tracer. The introduced tracking algorithm utilizes the pyramidal Lucas‐Kanade method and is especially suitable for thermal image data. The performance of the new tool is compared to traditional image‐based tracking tools, that is, PIVlab and PTVlab. Experiments are performed in the laboratory for three different flow velocities and tests are conducted in a small stream to illustrate the suitability of the tool for field measurements. Results of the laboratory experiments as well as of the field experiments show that our tracking algorithm, applied to imagery from a thermal camera, outperforms commonly used tracking methods. Our tool provides velocity fields with very high resolution and is in close agreement with reference measurements, whereas…

The Logarithmic Skew-Normal Distributions are Moment-Indeterminate

Gwo Dong Lin, Jordan Stoyanov

Statistics and Probability · Journal of Applied Probability · · cited 39× · DOI ↗ · more like this ·

We study the class of logarithmic skew-normal (LSN) distributions . They have heavy tails; however, all their moments of positive integer orders are finite. We are interested in the problem of moments for such distributions. We show that the LSN distributions are all nonunique (moment-indeterminate). Moreover, we explicitly describe Stieltjes classes for some LSN distributions; they are families of infinitely many distributions, which are different but have the same moment sequence as a fixed LSN distribution.

The Riemann Zeta Distribution

Gwo Dong Lin, Chin‐Yuan Hu

Algebra and Number Theory · Bernoulli · · cited 38× · DOI ↗ · more like this ·

Let ζ be the Riemann zeta function. Khinchine (1938) proved that the function f<sub>σ</sub>(t)=ζ(σ+i t)/ζ(σ), where σ> 1 and t is real, is an infinitely divisible characteristic function. We investigate further the fundamental properties of the corresponding distribution of f<sub>σ</sub>, the Riemann zeta distribution, including its support and unimodality. In particular, the Riemann zeta random variable is represented as a linear function of infinitely many independent geometric random variables. To extend Khinchine's result, we construct the Dirichlet-type characteristic functions of discrete distributions and provide a sufficient condition for the infinite divisibility of these characteristic functions. By way of applications, we give probabilistic proofs for some identities in number theory, including a new identity for the reciprocal of the Riemann zeta…

Hyper-BCube: A scalable data center network

Dong Lin, Liu Yang, Mounir Hamdi +1 more

Information Systems · Rare & Special e-Zone (The Hong Kong University of Science and Technology) · · cited 37× · DOI ↗ · more like this ·

Mega data centers are being built around the world to provide various cloud computing services. As a result, data center networking has recently been a hot research topic in both academia and industry. A fundamental challenge in this research is the design of the data center network that interconnects the massive number of servers, and provides efficient and fault-tolerant routing service to upper-layer applications. In response to this challenge, the research community have begun exploring novel interconnect topologies including Fat-Tree, DCell, and BCube and etc. Understandably, this research is still in its infancy. The proposed solutions either scale too fast (i.e., double exponentially) or too slow, suffer from performance bottlenecks, or can be quite costly in both routing and construction. In this paper, we propose a cost-effective and gracefully scalable data center interconnect…

LSP-based multiple-description coding for real-time low bit-rate voice over IP

Benjamin W. Wah, Dong Lin

Computer Vision and Pattern Recognition · IEEE Transactions on Multimedia · · cited 37× · DOI ↗ · more like this ·

A fundamental issue in real-time interactive voice transmissions over unreliable IF networks is the loss or late arrival of packets for playback. This problem is especially serious when transmitting low bit rate-coded speech with pervasive dependencies introduced. In this case, the loss or late arrival of a single packet will lead to the loss of subsequent dependent frames. We study end-to-end loss-concealment schemes for ensuring high quality in playback. We propose a novel multiple description-coding method for concealing packet losses in transmitting low bit rate-coded speech. Based on high correlations observed in linear predictor parameters-in the form of Line Spectral Pairs (LSPs)-of adjacent frames, we generate multiple descriptions in senders by interleaving LSPs, and reconstruct lost LSPs in receivers by linear interpolations. As excitation codewords have low correlations, we…

Use of Link-by-Link Flow Control in Maximizing ATM Network Performance: Simulation Results

H. T. Kung, Robert Morris, Thomas Charuhas +1 more

Computer Networks and Communications · · cited 36× · OpenAlex ↗ · more like this ·

Simulations have been performed to verify the effectiveness of using link-by-link flow controlled virtual channels for maximizing ATM network performance. A simulator which accurately reflects the real hardware design of a flow controlled ATM switch is used. The switch is currently under joint development by BNR and Harvard. The simulation results clearly demonstrate that the flow control mechanism is able to provide sufficiently rapid feedback to allow a network to adapt to load changes and maximize its performance. The simulations also show that, when compared to VCs using other traffic management approaches, flow controlled virtual circuits are efficient in terms of buffer usage and in point-to-multipoint multicast implementations. 1 Introduction A set of credit-based flow control schemes for implementing link-by-link flow controlled virtual channels has been proposed, in [5], for…

Scenario-based programming for SDN policies

Yifei Yuan, Dong Lin, Rajeev Alur +1 more

Computer Networks and Communications · · cited 35× · DOI ↗ · more like this ·

Recent emergence of software-defined networks offers an opportunity to design domain-specific programming abstractions aimed at network operators. In this paper, we propose scenario-based programming, a framework that allows network operators to program network policies by describing representative example behaviors. Given these scenarios, our synthesis algorithm automatically infers the controller state that needs to be maintained along with the rules to process network events and update state. We have developed the NetEgg scenario-based programming tool, which can execute the generated policy implementation on top of a centralized controller, but also automatically infers flow-table rules that can be pushed to switches to improve throughput. We study a range of policies considered in the literature and report our experience regarding specifying these policies using scenarios. We…