arXiv.itsnick.co

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

Author

Song Cheng

137 papers · 1994–2026

1994: 11995: 01996: 01997: 01998: 01999: 02000: 02001: 32002: 42003: 42004: 32005: 92006: 52007: 62008: 72009: 52010: 82011: 62012: 42013: 62014: 02015: 42016: 02017: 72018: 02019: 52020: 72021: 62022: 62023: 62024: 102025: 82026: 7
1994peak 10 in 20242026

Frequent co-authors

Papers

showing 1–25 of 137 RSS

Enhancing Item Response Theory for Cognitive Diagnosis

Song Cheng, Qi Liu

cs.AI · · cited 125× · arXiv ↗ · more like this ·

Cognitive diagnosis is a fundamental and crucial task in many educational applications, e.g., computer adaptive test and cognitive assignments. Item Response Theory (IRT) is a classical cognitive diagnosis method which can provide interpretable parameters (i.e., student latent trait, question discrimination, and difficulty) for analyzing student performance. However, traditional IRT ignores the rich information in question texts, cannot diagnose knowledge concept proficiency, and it is inaccurate to diagnose the parameters for the questions which only appear several times. To this end, in this paper, we propose a general Deep Item Response Theory (DIRT) framework to enhance traditional IRT for cognitive diagnosis by exploiting semantic representation from question texts with deep learning. In DIRT, we first use a proficiency vector to represent students' proficiency in knowledge…

Information Perspective to Probabilistic Modeling: Boltzmann Machines versus Born Machines

Song Cheng, Jing Chen, Lei Wang

physics.data-an · · cited 113× · arXiv ↗ · more like this ·

We compare and contrast the statistical physics and quantum physics inspired approaches for unsupervised generative modeling of classical data. The two approaches represent probabilities of observed data using energy-based models and quantum states respectively.Classical and quantum information patterns of the target datasets therefore provide principled guidelines for structural design and learning in these two approaches. Taking the restricted Boltzmann machines (RBM) as an example, we analyze the information theoretical bounds of the two approaches. We verify our reasonings by comparing the performance of RBMs of various architectures on the standard MNIST datasets.

Simulating Noisy Variational Quantum Algorithms: A Polynomial Approach

Yuguo Shao, Fuchuan Wei, Song Cheng +1 more

quant-ph · · cited 59× · arXiv ↗ · more like this ·

Large-scale variational quantum algorithms are widely recognized as a potential pathway to achieve practical quantum advantages. However, the presence of quantum noise might suppress and undermine these advantages, which blurs the boundaries of classical simulability. To gain further clarity on this matter, we present a novel polynomial-scale method based on the path integral of observable's back-propagation on Pauli paths (OBPPP). This method efficiently approximates expectation values of operators in variational quantum algorithms with bounded truncation error in the presence of single-qubit Pauli noise. Theoretically, we rigorously prove: 1) For a constant minimal non-zero noise rate $γ$, OBPPP's time and space complexity exhibit a polynomial relationship with the number of qubits $n$, the circuit depth $L$. 2) For variable $γ$, in scenarios where more than two non-zero noise factors…

Simulating Noisy Quantum Circuits with Matrix Product Density Operators

Song Cheng, Chenfeng Cao, Chao Zhang +4 more

quant-ph · · cited 54× · arXiv ↗ · more like this ·

Simulating quantum circuits with classical computers requires resources growing exponentially in terms of system size. Real quantum computer with noise, however, may be simulated polynomially with various methods considering different noise models. In this work, we simulate random quantum circuits in 1D with Matrix Product Density Operators (MPDO), for different noise models such as dephasing, depolarizing, and amplitude damping. We show that the method based on Matrix Product States (MPS) fails to approximate the noisy output quantum states for any of the noise models considered, while the MPDO method approximates them well. Compared with the method of Matrix Product Operators (MPO), the MPDO method reflects a clear physical picture of noise (with inner indices taking care of the noise simulation) and quantum entanglement (with bond indices taking care of two-qubit gate simulation).…

On Long-Range Dependence in NSFNET Traffic

Steven M. Klivansky, Amarnath Mukherjee, Song Cheng

Computer Networks and Communications · SMARTech Repository (Georgia Institute of Technology) · · cited 44× · OpenAlex ↗ · more like this ·

Several recent studies indicate that (i) dependent packet arrivals can have a significant impact on buffer-overflow probabilities and delay characteristics at a network multiplexor, with long-range dependence showing a larger impact than short-range dependence, and (ii) that traffic in several different campus-level network environments appears to be long-range dependent. These two related issues are beginning to make a significant impact on traffic theory and network resource management. The objective of this paper is to show that long-range dependence extends to wide-area networks as well, and to identify the factors contributing to it. In particular, we find the presence of long-range dependence in packet level traffic over the NSFNET wide-area network. The analysis is based on raw traffic data collected from multiple NSFNET core switches distributed across the US. Investigation of…

AdaptKT: A Domain Adaptable Method for Knowledge Tracing

Song Cheng, Qi Liu, Enhong Chen +5 more

Artificial Intelligence · Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining · · cited 23× · DOI ↗ · more like this ·

Knowledge tracing is a crucial and fundamental task in online education systems, which can predict students' knowledge state for personalized learning. Unfortunately, existing methods are domain-specific, whereas there are many domains (e.g., subjects, schools) in the real education scene and some domains suffer from the problem of lacking sufficient data. Therefore, how to exploit the knowledge in other domains, to improve the model's performance for target domain remains pretty much open. We term this problem as Domain Adaptation for Knowledge Tracing (DAKT), which aims to transfer knowledge from the source domain to the target one for knowledge tracing. In this paper, we propose a novel adaptable method, namely Adaptable Knowledge Tracing (AdaptKT), which contains three phases to explore this problem. Specifically, phase I is instance selection. Given the question texts of two…

An Intrusion Detection Method Based on Attention Mechanism to Improve CNN-BiLSTM Model

Dingyu Shou, Chao Li, Zhen Wang +5 more

Computer Networks and Communications · The Computer Journal · · cited 8× · DOI ↗ · more like this ·

Abstract Security of computer information can be improved with the use of a network intrusion detection system. Since the network environment is becoming more complex, more and more new methods of attacking the network have emerged, making the original intrusion detection methods ineffective. Increased network activity also causes intrusion detection systems to identify errors more frequently. We suggest a new intrusion detection technique in this research that combines a Convolutional Neural Network (CNN) model with a Bi-directional Long Short-term Memory Network (BiLSTM) model for adding attention mechanisms. We distinguish our model from existing methods in three ways. First, we use the NCR-SMOTE algorithm to resample the dataset. Secondly, we use recursive feature elimination method based on extreme random tree to select features. Thirdly, we improve the profitability and accuracy…

A High Speed SM3 Algorithm Implementation for Security Chip

Shiping Zang, Dongyan Zhao, Yi Hu +4 more

Hardware and Architecture · · cited 7× · DOI ↗ · more like this ·

High throughput of crypto circuit is critical for many high performance security applications. The proposed SM3 circuit design breaks the inherent limitation of the conventional algorithm flow by removing the "blocking point" on the critical path, and reorganizes the algebraic structure by adding four parallel compensation operations. In addition, the round expansion architecture, CSA (Carry Save Adder) and pre-calculation are also used in this design. Due to the optimization at both the algorithm level and the circuit level, the synthesized circuit of this design can reach maximum 415MHz operating clock frequency and 6.4Gbps throughput with SMIC 40nm high performance technology. Compared with the conventional implementation method, the throughput performance of the proposed SM3 circuit increases by 97.5% and the chip area of SM3 algorithm area is only increased by 16.2%.

Circular formation control for multi-agent systems with input saturation

Song Cheng, Yuan Fan

Computer Networks and Communications · · cited 7× · DOI ↗ · more like this ·

In this paper circular formation of a network of mobile agents with input saturation is considered. The goal is to drive the mobile agents to the configuration such that the distance between neighboring agents on a circle is consistent with prescribed distance. Distributed formation control laws with saturation constraints are developed which not only drive the agents to the desired configuration, but also preserve the mobile agents' order on the circle and thus collision avoidance among agents is guaranteed. A simulation example is also provided to illustrate the effectiveness of the proposed control scheme.

Blockchain-Based PKI System and Its Application in Internet of Things

Yanyan Yu, Yubo Wang, Yi Hu +5 more

Computer Networks and Communications · · cited 5× · DOI ↗ · more like this ·

With the continuous construction and development of the Internet of things, security and trust have become issues that cannot be ignored. Through transparent and trusted rules, blockchain can build a traceable chain data structure, which can effectively solve the problems of trust and security in the Internet of things. In this paper, a public key infrastructure suitable for the Internet of things is constructed, based on the characteristics of blockchain technology such as decentralization and tamper proof. The digital certificate is stored and verified in the way of blockchain, which can realize the decentralization of PKI, solve the problem of single point failure and trust problem of root CA, and improve the efficiency of certificate configuration.

Design of Lightweight AES Algorithm Based on Masked Lookup Table

Meng Li, Jie Gan, Song Cheng +3 more

Artificial Intelligence · · cited 5× · DOI ↗ · more like this ·

With the development of information and communication technology, block cipher algorithm represented by AES algorithm has been used more and more widely. This design optimizes the hardware implementation of AES algorithm, mainly including: using T lookup table to realize multiple operations in round unit, ameliorating the T lookup table by using fixed value masking to improve the safety, designing a structure of inner-round loop structure and completely expanded structure, and using ISE to synthesize the AES algorithm module on Virtex-6 platform. As a result of experiment, the inner-round loop structure takes up 410slices+136BRAM, and the completely expanded structure takes up 2492slices+732 BRAM. The consumption of slices increases by only 7%-10%, and can achieve 51.2Gbps throughput under 400MHz clock. Finally, the security analysis and DPA experiments show that the fixed value masking…

Zeno-free event-triggered consensus using sampled data

Yuan Fan, Wang Shi, Song Cheng

Computer Networks and Communications · · cited 5× · DOI ↗ · more like this ·

This work presents an event-triggered consensus algorithm for multi-agent systems with periodic sampling schemes. Combinational measurement error is employed to reduce the number of events. It is noted that by introducing the periodic sampling scheme, Zeno behaviors can be naturally avoided in both algorithms. The effectiveness of the proposed algorithms are illustrated by simulation examples.

Implementation of a Safe and Efficient Point Multiplication for SM2 Algorithm

Meng Li, Na Li, Huizhi Liu +3 more

Information Systems · · cited 4× · DOI ↗ · more like this ·

With the development of information security and communication technology, SM2 algorithm has been widely used in security chip, secure communication and other fields. Point multiplication is a key operation that affects the speed and security of SM2 algorithm. In order to make the implementation of SM2 algorithm safer and faster, this design optimized the fixed window method of implementation method of point multiplication, which mainly includes: mathematical transformation in the implementation process, adding a subtraction to achieve the purpose of security, and combining it with the point coordinate randomization, scalar randomization, and point verification security protection measures, achieving the effect of resisting common analyses against SM2 algorithm. Using the design compiler tool and UMC55 process library to synthesize the design, the module area is 29,793GE (equivalent…

Efficient Exploitation of Noise Leakage for Template Attack

Song Cheng, Hailong Zhang, Xiaobo Hu +2 more

Artificial Intelligence · IEEE Embedded Systems Letters · · cited 4× · DOI ↗ · more like this ·

Because of its strong leakage characterization ability and high-key-recovery efficiency, template attack (TA) is the strongest side-channel attack from the information-theoretic point-of-view. In light of this, TA can be used as a powerful tool to evaluate the security of a crypto device in real scenarios. In fact, by evaluating the key-recovery efficiency of TA, the security of a crypto device can be quantified. Therefore, it is very meaningful to optimize the key-recovery efficiency of TA in real scenarios. In this work, we show that the leakage exploitation style can significantly influence the key-recovery efficiency of TA. In detail, rather than exploiting the leakage of the noise part, the leakage of the squared noise part can be used in TA. Theoretically, by exploiting the leakage of the squared noise part, the key-recovery efficiency of TA can be optimized under the normalized…

Gradient Corrected Approximation for Binary Neural Networks

Song Cheng, Zixuan Li, Yongsen Wang +4 more

Computer Vision and Pattern Recognition · IEICE Transactions on Information and Systems · · cited 4× · DOI ↗ · more like this ·

Binary neural networks (BNNs), where both activations and weights are radically quantized to be {-1, +1}, can massively accelerate the run-time performance of convolution neural networks (CNNs) for edge devices, by computation complexity reduction and memory footprint saving. However, the non-differentiable binarizing function used in BNNs, makes the binarized models hard to be optimized, and introduces significant performance degradation than the full-precision models. Many previous works managed to correct the backward gradient of binarizing function with various improved versions of straight-through estimation (STE), or in a gradual approximate approach, but the gradient suppression problem was not analyzed and handled. Thus, we propose a novel gradient corrected approximation (GCA) method to match the discrepancy between binarizing function and backward gradient in a gradual and…

Hardware Implementation and optimization Design of Lightweight RECTANGLE Algorithm

Meng Li, Dongyan Zhao, Xiaoke Tang +3 more

Artificial Intelligence · · cited 4× · DOI ↗ · more like this ·

With the development of information security, RFID and other fields, lightweight algorithms have been widely used. In order to make the lightweight algorithm applied to more situations, this design optimizes the hardware implementation of the lightweight RECTANGLE algorithm, which mainly includes: solving the Boolean expression of S-box, deriving the inverse key schedule in decryption, designing the cyclic iterative structure. Finally, under UMC55 process, the areas of REC-80 and REC-128 encryption modules are 2152. 8GE and 2916GE, which are 13.16% and 8.38% less than before optimization. And the throughput can reach 980Mbps under 400MHz. The areas of REC-80 and REC-128 encryption/decryption modules are 3336. 8GE and 4866. 8GE under the inverse key schedule mode. And the throughput can reach 820Mbps under 333MHz. Compared with before, these modules have obvious optimization effect…

Optimized Design and Implementation of CRYSTALS‐KYBER Based on MLWE

Song Cheng, J.-Y. Chen, Jianyang Li +4 more

Artificial Intelligence · Security and Communication Networks · · cited 3× · DOI ↗ · more like this ·

CRYSTALS‐KYBER is a postquantum key exchange mechanism based on the MLWE problem. At present, it is a major challenge to realize CRYSTALS‐KYBER with high‐performance and low‐resource consumption on the FPGA platform. The design proposed in this paper aims at the high efficiency implementation of CRYSTALS‐KYBER based on the MLWE scheme. In this work, multipath delay commutator–number theoretic transform (MDC‐NTT) and area‐optimized hash unit are employed to optimize the overhead and clock cycle occupancy of the whole implementation. The CRYSTALS‐KYBER public‐key encapsulation circuit requires only 5551 clock cycles to execute, reducing the latency by 38.9% compared to the best existing iteration NTT designs and improving hardware efficiency by 1.6–1.9 times. The maximum operating frequency also reaches up to 225 MHz.

AMA-Det: Enhancing Shared Head of One-Stage Object Detection With Adaptation, Merging, and Alignment

Song Cheng, Feng-Yue Li, Shushan Qiao +2 more

Computer Vision and Pattern Recognition · IEEE Access · · cited 3× · DOI ↗ · more like this ·

Feature adaptation(FA) and result alignment(RA) are critical issues in one-stage object detection, since FA amends feature misalignment problem by adapting sampling points to semantically significant locations, and RA corrects result misalignment problem by estimating the localization quality. For FA, the aligned and consistent prediction of sampling points is important. Previous studies directly inherit cascaded “proposal-refine” philosophy from multi-stage detectors and predict sampling points by approximating their minimal external rectangle to ground-truth bounding boxes. This manner generates poorly aligned and consistent sampling points and induces irrelevant features aggregated. Moreover, their sampling points for classification are conventionally generated through the localization branch, without feeding the corresponding features of the classification branch. For RA, previous…

Increasing Information Entropy of Both Weights and Activations for the Binary Neural Networks

Wanbing Zou, Song Cheng, Luyuan Wang +4 more

Computer Vision and Pattern Recognition · Electronics · · cited 3× · DOI ↗ · more like this ·

In terms of memory footprint requirement and computing speed, the binary neural networks (BNNs) have great advantages in power-aware deployment applications, such as AIoT edge terminals, wearable and portable devices, etc. However, the networks’ binarization process inevitably brings considerable information losses, and further leads to accuracy deterioration. To tackle these problems, we initiate analyzing from a perspective of the information theory, and manage to improve the networks information capacity. Based on the analyses, our work has two primary contributions: the first is a newly proposed median loss (ML) regularization technique. It improves the binary weights distribution more evenly, and consequently increases the information capacity of BNNs greatly. The second is the batch median of activations (BMA) method. It raises the entropy of activations by subtracting a median…

Tree Tensor Networks for Generative Modeling

Song Cheng, Lei Wang, Tao Xiang +1 more

stat.ML · · cited 3× · arXiv ↗ · more like this ·

Matrix product states (MPS), a tensor network designed for one-dimensional quantum systems, has been recently proposed for generative modeling of natural data (such as images) in terms of `Born machine'. However, the exponential decay of correlation in MPS restricts its representation power heavily for modeling complex data such as natural images. In this work, we push forward the effort of applying tensor networks to machine learning by employing the Tree Tensor Network (TTN) which exhibits balanced performance in expressibility and efficient training and sampling. We design the tree tensor network to utilize the 2-dimensional prior of the natural images and develop sweeping learning and sampling algorithms which can be efficiently implemented utilizing Graphical Processing Units (GPU). We apply our model to random binary patterns and the binary MNIST datasets of handwritten digits. We…

Potential propensity of traditional herbal materials ingested by collegiate athletes in South China for their consuming health care

Yue-Quan Qin, Chun-Song Cheng, Ying Jiang +3 more

Sociology and Political Science · Medicine · · cited 2× · DOI ↗ · more like this ·

ABSTRACT: In south China, traditional herbal medicines have been widely used as functional foods or dietary supplements for daily health care. Many plant-derived chemical substances with biological activity are inadvertently ingested by collegiate athletes daily through canton-style herbal tea or herbal slow-cooked soup. In the view of the complexity of herbal ingredients, it is still no full survey reported for the sports risk of plant-derived sports doping. This research is firstly a descriptive statistical analysis. Collegiate athletes with different socio-economic characteristics from medical colleges in 3 different regions in China participated in the questionnaire survey. Three survey forms, including the oral interview, email inquiry, handing out and recovering the questionnaires in live, were developed and performed by researchers. It was first found that collegiate athletes…

Domain Adaption for Knowledge Tracing

Song Cheng, Qi Liu, Enhong Chen

cs.LG · · cited 2× · arXiv ↗ · more like this ·

With the rapid development of online education system, knowledge tracing which aims at predicting students' knowledge state is becoming a critical and fundamental task in personalized education. Traditionally, existing methods are domain-specified. However, there are a larger number of domains (e.g., subjects, schools) in the real world and the lacking of data in some domains, how to utilize the knowledge and information in other domains to help train a knowledge tracing model for target domains is increasingly important. We refer to this problem as domain adaptation for knowledge tracing (DAKT) which contains two aspects: (1) how to achieve great knowledge tracing performance in each domain. (2) how to transfer good performed knowledge tracing model between domains. To this end, in this paper, we propose a novel adaptable framework, namely adaptable knowledge tracing (AKT) to address…

Resident Trip Characteristics Comparison in Main Cities Among Three City Circles of China

Song Cheng

Transportation · Traffic & Transportation · · cited 2× · OpenAlex ↗ · more like this ·

As the most developed economic regions,traffic problems in main cities of three city circles are very serious.Eleven main cities are chosen in this paper.It analyses the correlation characteristics in the average number of resident trip,trip purpose,trip mode and trip time consume of both cities,and also seeks the internal relations between resident trip characteristics and city economy and size,which could lay a foundation for alleviating traffic problems.

Architecture and Implementation of Service Management Based on Active Network and CORBA

Song Cheng

Information Systems · · cited 2× · OpenAlex ↗ · more like this ·

This paper presents a new concept of service component,establishes the service management architecture based on active network and CORBA.With the help of active network and CORBA,the service management system can improve its efficiency remarkably and meet the needs of distributed and heterogeneous environments.Analyzes the component,function and character of the architecture,and presents a prototype.

Holographic Bit Threads from String-Diagrammatic Quantum Information Flow

Yi-Yu Lin, Song Cheng

hep-th · · cited 1× · arXiv ↗ · more like this ·

Bit threads, arising as the convex dual of the minimal-surface formula for holographic entanglement entropy, are line-like structures with nontrivial bulk trajectories. Their physical picture has long been associated with Bell-pair interpretations, yet the meaning of their detailed trajectories, particularly their nonuniqueness, remains unclear. We propose to apply Coecke's notion of quantum information flow (QIF) in protocol-network geometry, developed within categorical quantum mechanics, to a holographic setup, and show that its trajectories within the discrete holographic bulk---the tensor network---obey the divergence-free and density-bound conditions of bit threads. From this process-centered perspective, bit-thread nonuniqueness becomes natural: for a given resource state, different protocols realizing the same distillation task can give rise to different QIF trajectories. We…