特征增强的Sparse Transformer目标跟踪算法

针对Transformer的自注意力机制计算量大、容易被背景分心,导致有效信息抓取不足,从而降低跟踪性能的问题,提出特征增强的Sparse Transformer目标跟踪算法。基于孪生网络骨干进行特征提取;特征增强模块利用多尺度特征图生成的上下文信息,增强目标局部特征;利用Sparse Transformer的最相关特性生成目标聚焦特征,并嵌入位置编码提升跟踪定位的精度。提出的跟踪模型以端到端的方式进行训练,在OTB100,VOT2018和LaSOT等5个数据集上进行了大量实验,实验结果表明所提算法取得了较好的跟踪性能,实时跟踪速度为34帧/s。

高斯线性模型正则估计的Cramér-Rao下界

该文针对正则化高斯模型中参数估计的Cramér-Rao下界(CRB)开展研究,提出了一种新型CRB.在线性高斯模型的设计矩阵单位正交的假设下,给出L_1类正则估计的方差及CRB的显式表达,并进行数值计算.进一步地,推导了正则估计的CRB取等条件:在线性高斯模型中,取得CRB的估计均为线性估计量;在正则项可微的假设下,仅二次多项式正则项可令估计取得CRB.最后,针对带稀疏特征的估计提出sparse CRB,将其与现有的CRB比较,从理论和实践两方面说明了其优势.

2D DOA Estimation of Coherent Signals with a Separated Linear Acoustic Vector-Sensor Array

In this paper, a two-dimensional(2D) DOA estimation algorithm of coherent signals with a separated linear acoustic vector-sensor(AVS) array consisting of two sparse AVS arrays is proposed. Firstly,the partitioned spatial smoothing(PSS) technique is used to construct a block covariance matrix, so as to decorrelate the coherency of signals. Then a signal subspace can be obtained by singular value decomposition(SVD) of the covariance matrix. Using the signal subspace, two extended signal subspaces are constructed to compensate aperture loss caused by PSS.The elevation angles can be estimated by estimation of signal parameter via rotational invariance techniques(ESPRIT) algorithm. At last, the estimated elevation angles can be used to estimate automatically paired azimuth angles. Compared with some other ESPRIT algorithms, the proposed algorithm shows higher estimation accuracy, which can be proved through the simulation results.

一种基于迭代近端投影的被动声纳探测离网格DOA估计方法

Traditional direction of arrival(DOA)estimation methods based on sparse reconstruction commonly use convex or smooth functions to approximate non-convex and non-smooth sparse representation problems.This approach often introduces errors into the sparse representation model,necessitating the development of improved DOA estimation algorithms.Moreover,conventional DOA estimation methods typically assume that the signal coincides with a predetermined grid.However,in reality,this assumption often does not hold true.The likelihood of a signal not aligning precisely with the predefined grid is high,resulting in potential grid mismatch issues for the algorithm.To address the challenges associated with grid mismatch and errors in sparse representation models,this article proposes a novel high-performance off-grid DOA estimation approach based on iterative proximal projection(IPP).In the proposed method,we employ an alternating optimization strategy to jointly estimate sparse signals and grid offset parameters.A proximal function optimization model is utilized to address non-convex and non-smooth sparse representation problems in DOA estimation.Subsequently,we leverage the smoothly clipped absolute deviation penalty(SCAD)function to compute the proximal operator for solving the model.Simulation and sea trial experiments have validated the superiority of the proposed method in terms of higher resolution and more accurate DOA estimation performance when compared to both traditional sparse reconstruction methods and advanced off-grid techniques.

Sparse Reconstructive Evidential Clustering for Multi-View Data

Although many multi-view clustering(MVC) algorithms with acceptable performances have been presented, to the best of our knowledge, nearly all of them need to be fed with the correct number of clusters. In addition, these existing algorithms create only the hard and fuzzy partitions for multi-view objects,which are often located in highly-overlapping areas of multi-view feature space. The adoption of hard and fuzzy partition ignores the ambiguity and uncertainty in the assignment of objects, likely leading to performance degradation. To address these issues, we propose a novel sparse reconstructive multi-view evidential clustering algorithm(SRMVEC). Based on a sparse reconstructive procedure, SRMVEC learns a shared affinity matrix across views, and maps multi-view objects to a 2-dimensional humanreadable chart by calculating 2 newly defined mathematical metrics for each object. From this chart, users can detect the number of clusters and select several objects existing in the dataset as cluster centers. Then, SRMVEC derives a credal partition under the framework of evidence theory, improving the fault tolerance of clustering. Ablation studies show the benefits of adopting the sparse reconstructive procedure and evidence theory. Besides,SRMVEC delivers effectiveness on benchmark datasets by outperforming some state-of-the-art methods.

基于吉布斯采样的稀疏水声信道估计方法

The estimation of sparse underwater acoustic(UWA)channels can be regarded as an inference problem involving hidden variables within the Bayesian framework.While the classical sparse Bayesian learning(SBL),derived through the expectation maximization(EM)algorithm,has been widely employed for UWA channel estimation,it still differs from the real posterior expectation of channels.In this paper,we propose an approach that combines variational inference(VI)and Markov chain Monte Carlo(MCMC)methods to provide a more accurate posterior estimation.Specifically,the SBL is first re-derived with VI,allowing us to replace the posterior distribution of the hidden variables with a variational distribution.Then,we determine the full conditional probability distribution for each variable in the variational distribution and then iteratively perform random Gibbs sampling in MCMC to converge the Markov chain.The results of simulation and experiment indicate that our estimation method achieves lower mean square error and bit error rate compared to the classic SBL approach.Additionally,it demonstrates an acceptable convergence speed.

THE SPARSE REPRESENTATION RELATED WITH FRACTIONAL HEAT EQUATIONS

This study introduces a pre-orthogonal adaptive Fourier decomposition(POAFD)to obtain approximations and numerical solutions to the fractional Laplacian initial value problem and the extension problem of Caffarelli and Silvestre(generalized Poisson equation).As a first step,the method expands the initial data function into a sparse series of the fundamental solutions with fast convergence,and,as a second step,makes use of the semigroup or the reproducing kernel property of each of the expanding entries.Experiments show the effectiveness and efficiency of the proposed series solutions.

A Novel Clutter Suppression Algorithm for Low-Slow-Small Targets Detecting Based on Sparse Adaptive Filtering

Passive detection of low-slow-small(LSS)targets is easily interfered by direct signal and multipath clutter,and the traditional clutter suppression method has the contradiction between step size and convergence rate.In this paper,a frequency domain clutter suppression algorithm based on sparse adaptive filtering is proposed.The pulse compression operation between the error signal and the input reference signal is added to the cost function as a sparsity constraint,and the criterion for filter weight updating is improved to obtain a purer echo signal.At the same time,the step size and penalty factor are brought into the adaptive iteration process,and the input data is used to drive the adaptive changes of parameters such as step size.The proposed algorithm has a small amount of calculation,which improves the robustness to parameters such as step size,reduces the weight error of the filter and has a good clutter suppression performance.

A Time Series Intrusion Detection Method Based on SSAE,TCN and Bi-LSTM

In the fast-evolving landscape of digital networks,the incidence of network intrusions has escalated alarmingly.Simultaneously,the crucial role of time series data in intrusion detection remains largely underappreciated,with most systems failing to capture the time-bound nuances of network traffic.This leads to compromised detection accuracy and overlooked temporal patterns.Addressing this gap,we introduce a novel SSAE-TCN-BiLSTM(STL)model that integrates time series analysis,significantly enhancing detection capabilities.Our approach reduces feature dimensionalitywith a Stacked Sparse Autoencoder(SSAE)and extracts temporally relevant features through a Temporal Convolutional Network(TCN)and Bidirectional Long Short-term Memory Network(Bi-LSTM).By meticulously adjusting time steps,we underscore the significance of temporal data in bolstering detection accuracy.On the UNSW-NB15 dataset,ourmodel achieved an F1-score of 99.49%,Accuracy of 99.43%,Precision of 99.38%,Recall of 99.60%,and an inference time of 4.24 s.For the CICDS2017 dataset,we recorded an F1-score of 99.53%,Accuracy of 99.62%,Precision of 99.27%,Recall of 99.79%,and an inference time of 5.72 s.These findings not only confirm the STL model’s superior performance but also its operational efficiency,underpinning its significance in real-world cybersecurity scenarios where rapid response is paramount.Our contribution represents a significant advance in cybersecurity,proposing a model that excels in accuracy and adaptability to the dynamic nature of network traffic,setting a new benchmark for intrusion detection systems.

Triad-displaced ULAs configuration for non-circular sources with larger continuous virtual aperture and enhanced degrees of freedom

Non-uniform linear array(NULA)configurations are well renowned due to their structural ability for providing increased degrees of freedom(DOF)and wider array aperture than uniform linear arrays(ULAs).These characteristics play a significant role in improving the direction-of-arrival(DOA)estimation accuracy.However,most of the existing NULA geometries are primarily applicable to circular sources(CSs),while they limitedly improve the DOF and continuous virtual aperture for noncircular sources(NCSs).Toward this purpose,we present a triaddisplaced ULAs(Tdis-ULAs)configuration for NCS.The TdisULAs structure generally consists of three ULAs,which are appropriately placed.The proposed antenna array approach fully exploits the non-circular characteristics of the sources.Given the same number of elements,the Tdis-ULAs design achieves more DOF and larger hole-free co-array aperture than its sparse array competitors.Advantageously,the number of uniform DOF,optimal distribution of elements among the ULAs,and precise element positions are uniquely determined by the closed-form expressions.Moreover,the proposed array also produces a filled resulting co-array.Numerical simulations are conducted to show the performance advantages of the proposed Tdis-ULAs configuration over its counterpart designs.

Monitoring seismicity in the southern Sichuan Basin using a machine learning workflow

Monitoring seismicity in real time provides significant benefits for timely earthquake warning and analyses.In this study,we propose an automatic workflow based on machine learning(ML)to monitor seismicity in the southern Sichuan Basin of China.This workflow includes coherent event detection,phase picking,and earthquake location using three-component data from a seismic network.By combining Phase Net,we develop an ML-based earthquake location model called Phase Loc,to conduct real-time monitoring of the local seismicity.The approach allows us to use synthetic samples covering the entire study area to train Phase Loc,addressing the problems of insufficient data samples,imbalanced data distribution,and unreliable labels when training with observed data.We apply the trained model to observed data recorded in the southern Sichuan Basin,China,between September 2018 and March 2019.The results show that the average differences in latitude,longitude,and depth are 5.7 km,6.1 km,and 2 km,respectively,compared to the reference catalog.Phase Loc combines all available phase information to make fast and reliable predictions,even if only a few phases are detected and picked.The proposed workflow may help real-time seismic monitoring in other regions as well.

An Efficient Radar Detection Method of Maneuvering Small Targets

Detection of maneuvering small targets has always been an important yet challenging task for radar signal processing.One primary reason is that target variable motions within coherent processing interval generate energy migrations across multiple resolution bins,which severely deteriorate the parameter estimation performance.A coarse-to-fine strategy for the detection of maneuvering small targets is proposed.Integration of small points segmented coherently is performed first,and then an optimal inter-segment integration is utilized to derive the coarse estimation of the chirp rate.Sparse fractional Fourier transform(FrFT)is then employed to refine the coarse estimation at a significantly reduced computational complexity.Simulation results verify the proposed scheme that achieves an efficient and reliable maneuvering target detection with-16dB input signal-to-noise ratio(SNR),while requires no exact a priori knowledge on the motion parameters.

Cambricon-QR:a sparse and bitwise reproducible quantized training accelerator

Quantized training has been proven to be a prominent method to achieve deep neural network training under limited computational resources.It uses low bit-width arithmetics with a proper scaling factor to achieve negligible accuracy loss.Cambricon-Q is the ASIC design proposed to efficiently support quantized training,and achieves significant performance improvement.However,there are still two caveats in the design.First,Cambricon-Q with different hardware specifications may lead to different numerical errors,resulting in non-reproducible behaviors which may become a major concern in critical applications.Second,Cambricon-Q cannot leverage data sparsity,where considerable cycles could still be squeezed out.To address the caveats,the acceleration core of Cambricon-Q is redesigned to support fine-grained irregular data processing.The new design not only enables acceleration on sparse data,but also enables performing local dynamic quantization by contiguous value ranges(which is hardware independent),instead of contiguous addresses(which is dependent on hardware factors).Experimental results show that the accuracy loss of the method still keeps negligible,and the accelerator achieves 1.61×performance improvement over Cambricon-Q,with about 10%energy increase.

Design Framework of Unsourced Multiple Access for 6G Massive IoT

In this paper,ambient IoT is used as a typical use case of massive connections for the sixth generation(6G)mobile communications where we derive the performance requirements to facilitate the evaluation of technical solutions.A rather complete design of unsourced multiple access is proposed in which two key parts:a compressed sensing module for active user detection,and a sparse interleaver-division multiple access(SIDMA)module are simulated side by side on a same platform at balanced signal to noise ratio(SNR)operating points.With a proper combination of compressed sensing matrix,a convolutional encoder,receiver algorithms,the simulated performance results appear superior to the state-of-the-art benchmark,yet with relatively less complicated processing.

Robustness Assessment of Asynchronous Advantage Actor-Critic Based on Dynamic Skewness and Sparseness Computation: A Parallel Computing View

Reinforcement learning as autonomous learning is greatly driving artificial intelligence(AI)development to practical applications.Having demonstrated the potential to significantly improve synchronously parallel learning,the parallel computing based asynchronous advantage actor-critic(A3C)opens a new door for reinforcement learning.Unfortunately,the acceleration's influence on A3C robustness has been largely overlooked.In this paper,we perform the first robustness assessment of A3C based on parallel computing.By perceiving the policy's action,we construct a global matrix of action probability deviation and define two novel measures of skewness and sparseness to form an integral robustness measure.Based on such static assessment,we then develop a dynamic robustness assessing algorithm through situational whole-space state sampling of changing episodes.Extensive experiments with different combinations of agent number and learning rate are implemented on an A3C-based pathfinding application,demonstrating that our proposed robustness assessment can effectively measure the robustness of A3C,which can achieve an accuracy of 83.3%.

Estimating High-Order Functional Connectivity Networks for Mild Cognitive Impairment Identification Based on Topological Structure

Functional connectivity networks (FCNs) are important in the diagnosis of neurological diseases and the understanding of brain tissue patterns. Recently, many methods, such as Pearson’s correlation (PC), Sparse representation (SR), and Sparse low-rank representation have been proposed to estimate FCNs. Despite their popularity, they only capture the low-order connections of the brain regions, failing to encode more complex relationships (i.e. , high-order relationships). Although researchers have proposed high-order methods, like PC + PC and SR + SR, aiming to build FCNs that can reflect more real state of the brain. However, such methods only consider the relationships between brain regions during the FCN construction process, neglecting the potential shared topological structure information between FCNs of different subjects. In addition, the low-order relationships are always neglected during the construction of high-order FCNs. To address these issues, in this paper we proposed a novel method, namely Ho-FCNTops, towards estimating high-order FCNs based on brain topological structure. Specifically, inspired by the Group-constrained sparse representation (GSR), we first introduced a prior assumption that all subjects share the same topological structure in the construction of the low-order FCNs. Subsequently, we employed the Correlation-reserved embedding (COPE) to eliminate noise and redundancy from the low-order FCNs. Meanwhile, we retained the original low-order relationships during the embedding process to obtain new node representations. Finally, we utilized the SR method on the obtained new node representations to construct the Ho-FCNTops required for disease identification. To validate the effectiveness of the proposed method, experiments were conducted on 137 subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database to identify Mild Cognitive Impairment (MCI) patients from the normal controls. The experimental results demonstrate superior performance compared to baseline methods.

一种轻量化三维人体姿态估计算法

针对三维人体姿态估计实际应用场景需求,提出一种基于空洞卷积ResNet模块和稀疏自注意力(Sparse Attention,SA)的轻量化三维人体姿态估计模型DS-Net(Dilated Sparse Attention Network)。首先,以单目、单阶段、多个三维人的回归网络(Monocular,One-stage,Regression of Multiple 3D People,ROMP)为基础姿态估计模型,并替换支路中基础ResNet模块的卷积为空洞卷积,在不降低精度的前提下减少模型参数量;其次,在支路中嵌入Sparse Attention,加强上下文理解能力以提高精度;最后,经过7个数据集训练和3DPW数据集测试,验证模型可行性。经实验验证,提出的DS-Net总参数量减少53.8%;在三维人体姿态估计任务中与ROMP相比,MPJPE和PA-MPJPE分别降低1.8%和2.9%,满足姿态估计实际应用场景需求。

Low-Rank Multi-View Subspace Clustering Based on Sparse Regularization

Multi-view Subspace Clustering (MVSC) emerges as an advanced clustering method, designed to integrate diverse views to uncover a common subspace, enhancing the accuracy and robustness of clustering results. The significance of low-rank prior in MVSC is emphasized, highlighting its role in capturing the global data structure across views for improved performance. However, it faces challenges with outlier sensitivity due to its reliance on the Frobenius norm for error measurement. Addressing this, our paper proposes a Low-Rank Multi-view Subspace Clustering Based on Sparse Regularization (LMVSC- Sparse) approach. Sparse regularization helps in selecting the most relevant features or views for clustering while ignoring irrelevant or noisy ones. This leads to a more efficient and effective representation of the data, improving the clustering accuracy and robustness, especially in the presence of outliers or noisy data. By incorporating sparse regularization, LMVSC-Sparse can effectively handle outlier sensitivity, which is a common challenge in traditional MVSC methods relying solely on low-rank priors. Then Alternating Direction Method of Multipliers (ADMM) algorithm is employed to solve the proposed optimization problems. Our comprehensive experiments demonstrate the efficiency and effectiveness of LMVSC-Sparse, offering a robust alternative to traditional MVSC methods.

Robust Principal Component Analysis Integrating Sparse and Low-Rank Priors

Principal Component Analysis (PCA) is a widely used technique for data analysis and dimensionality reduction, but its sensitivity to feature scale and outliers limits its applicability. Robust Principal Component Analysis (RPCA) addresses these limitations by decomposing data into a low-rank matrix capturing the underlying structure and a sparse matrix identifying outliers, enhancing robustness against noise and outliers. This paper introduces a novel RPCA variant, Robust PCA Integrating Sparse and Low-rank Priors (RPCA-SL). Each prior targets a specific aspect of the data’s underlying structure and their combination allows for a more nuanced and accurate separation of the main data components from outliers and noise. Then RPCA-SL is solved by employing a proximal gradient algorithm for improved anomaly detection and data decomposition. Experimental results on simulation and real data demonstrate significant advancements.

A Fast Clustering Based Evolutionary Algorithm for Super-Large-Scale Sparse Multi-Objective Optimization

During the last three decades,evolutionary algorithms(EAs)have shown superiority in solving complex optimization problems,especially those with multiple objectives and non-differentiable landscapes.However,due to the stochastic search strategies,the performance of most EAs deteriorates drastically when handling a large number of decision variables.To tackle the curse of dimensionality,this work proposes an efficient EA for solving super-large-scale multi-objective optimization problems with sparse optimal solutions.The proposed algorithm estimates the sparse distribution of optimal solutions by optimizing a binary vector for each solution,and provides a fast clustering method to highly reduce the dimensionality of the search space.More importantly,all the operations related to the decision variables only contain several matrix calculations,which can be directly accelerated by GPUs.While existing EAs are capable of handling fewer than 10000 real variables,the proposed algorithm is verified to be effective in handling 1000000 real variables.Furthermore,since the proposed algorithm handles the large number of variables via accelerated matrix calculations,its runtime can be reduced to less than 10%of the runtime of existing EAs.