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。

A Two-Layer Encoding Learning Swarm Optimizer Based on Frequent Itemsets for Sparse Large-Scale Multi-Objective Optimization

Traditional large-scale multi-objective optimization algorithms(LSMOEAs)encounter difficulties when dealing with sparse large-scale multi-objective optimization problems(SLM-OPs)where most decision variables are zero.As a result,many algorithms use a two-layer encoding approach to optimize binary variable Mask and real variable Dec separately.Nevertheless,existing optimizers often focus on locating non-zero variable posi-tions to optimize the binary variables Mask.However,approxi-mating the sparse distribution of real Pareto optimal solutions does not necessarily mean that the objective function is optimized.In data mining,it is common to mine frequent itemsets appear-ing together in a dataset to reveal the correlation between data.Inspired by this,we propose a novel two-layer encoding learning swarm optimizer based on frequent itemsets(TELSO)to address these SLMOPs.TELSO mined the frequent terms of multiple particles with better target values to find mask combinations that can obtain better objective values for fast convergence.Experi-mental results on five real-world problems and eight benchmark sets demonstrate that TELSO outperforms existing state-of-the-art sparse large-scale multi-objective evolutionary algorithms(SLMOEAs)in terms of performance and convergence speed.

Enhancing Evolutionary Algorithms With Pattern Mining for Sparse Large-Scale Multi-Objective Optimization Problems

Sparse large-scale multi-objective optimization problems(SLMOPs)are common in science and engineering.However,the large-scale problem represents the high dimensionality of the decision space,requiring algorithms to traverse vast expanse with limited computational resources.Furthermore,in the context of sparse,most variables in Pareto optimal solutions are zero,making it difficult for algorithms to identify non-zero variables efficiently.This paper is dedicated to addressing the challenges posed by SLMOPs.To start,we introduce innovative objective functions customized to mine maximum and minimum candidate sets.This substantial enhancement dramatically improves the efficacy of frequent pattern mining.In this way,selecting candidate sets is no longer based on the quantity of nonzero variables they contain but on a higher proportion of nonzero variables within specific dimensions.Additionally,we unveil a novel approach to association rule mining,which delves into the intricate relationships between non-zero variables.This novel methodology aids in identifying sparse distributions that can potentially expedite reductions in the objective function value.We extensively tested our algorithm across eight benchmark problems and four real-world SLMOPs.The results demonstrate that our approach achieves competitive solutions across various challenges.

THE STABLE RECONSTRUCTION OF STRONGLY-DECAYING BLOCK SPARSE SIGNALS

In this paper,we reconstruct strongly-decaying block sparse signals by the block generalized orthogonal matching pursuit(BgOMP)algorithm in the l2-bounded noise case.Under some restraints on the minimum magnitude of the nonzero elements of the strongly-decaying block sparse signal,if the sensing matrix satisfies the the block restricted isometry property(block-RIP),then arbitrary strongly-decaying block sparse signals can be accurately and steadily reconstructed by the BgOMP algorithm in iterations.Furthermore,we conjecture that this condition is sharp.

Progress on two-dimensional ferrovalley materials

The electron's charge and spin degrees of freedom are at the core of modern electronic devices. With the in-depth investigation of two-dimensional materials, another degree of freedom, valley, has also attracted tremendous research interest. The intrinsic spontaneous valley polarization in two-dimensional magnetic systems, ferrovalley material, provides convenience for detecting and modulating the valley. In this review, we first introduce the development of valleytronics.Then, the valley polarization forms by the p-, d-, and f-orbit that are discussed. Following, we discuss the investigation progress of modulating the valley polarization of two-dimensional ferrovalley materials by multiple physical fields, such as electric, stacking mode, strain, and interface. Finally, we look forward to the future developments of valleytronics.

Recent advances in two-dimensional photovoltaic devices

Two-dimensional(2D)materials have attracted tremendous interest in view of the outstanding optoelectronic properties,showing new possibilities for future photovoltaic devices toward high performance,high specific power and flexibility.In recent years,substantial works have focused on 2D photovoltaic devices,and great progress has been achieved.Here,we present the review of recent advances in 2D photovoltaic devices,focusing on 2D-material-based Schottky junctions,homojunctions,2D−2D heterojunctions,2D−3D heterojunctions,and bulk photovoltaic effect devices.Furthermore,advanced strategies for improving the photovoltaic performances are demonstrated in detail.Finally,conclusions and outlooks are delivered,providing a guideline for the further development of 2D photovoltaic devices.

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.

Alternating spin splitting of electronic and magnon bands in two-dimensional altermagnetic materials

Unconventional antiferromagnetism dubbed as altermagnetism was first discovered in rutile structured magnets,which is featured by spin splitting even without the spin–orbital coupling effect.This interesting phenomenon has been discovered in more altermagnetic materials.In this work,we explore two-dimensional altermagnetic materials by studying two series of two-dimensional magnets,including MF4 with M covering all 3d and 4d transition metal elements,as well as TS2 with T=V,Cr,Mn,Fe.Through the magnetic symmetry operation of RuF4 and MnS2,it is verified that breaking the time inversion is a necessary condition for spin splitting.Based on symmetry analysis and first-principles calculations,we find that the electronic bands and magnon dispersion experience alternating spin splitting along the same path.This work paves the way for exploring altermagnetism in two-dimensional materials.

Unlocking the potential of ultra-thin two-dimensional antimony materials:Selective growth and carbon coating for efficient potassium-ion storage

Antimony-based anodes have attracted wide attention in potassium-ion batteries due to their high theoretical specific capacities(∼660 mA h g^(-1))and suitable voltage platforms.However,severe capacity fading caused by huge volume change and limited ion transportation hinders their practical applications.Recently,strategies for controlling the morphologies of Sb-based materials to improve the electrochemical performances have been proposed.Among these,the two-dimensional Sb(2D-Sb)materials present excellent properties due to shorted ion immigration paths and enhanced ion diffusion.Nevertheless,the synthetic methods are usually tedious,and even the mechanism of these strategies remains elusive,especially how to obtain large-scale 2D-Sb materials.Herein,a novel strategy to synthesize 2D-Sb material using a straightforward solvothermal method without the requirement of a complex nanostructure design is provided.This method leverages the selective adsorption of aldehyde groups in furfural to induce crystal growth,while concurrently reducing and coating a nitrogen-doped carbon layer.Compared to the reported methods,it is simpler,more efficient,and conducive to the production of composite nanosheets with uniform thickness(3–4 nm).The 2D-Sb@NC nanosheet anode delivers an extremely high capacity of 504.5 mA h g^(-1) at current densities of 100 mA g^(-1) and remains stable for more than 200 cycles.Through characterizations and molecular dynamic simulations,how potassium storage kinetics between 2D Sb-based materials and bulk Sb-based materials are explored,and detailed explanations are provided.These findings offer novel insights into the development of durable 2D alloy-based anodes for next-generation potassium-ion batteries.

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.

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.

Recent progress on valley polarization and valley-polarized topological states in two-dimensional materials

Valleytronics, using valley degree of freedom to encode, process, and store information, may find practical applications in low-power-consumption devices. Recent theoretical and experimental studies have demonstrated that twodimensional(2D) honeycomb lattice systems with inversion symmetry breaking, such as transition-metal dichalcogenides(TMDs), are ideal candidates for realizing valley polarization. In addition to the optical field, lifting the valley degeneracy of TMDs by introducing magnetism is an efficient way to manipulate the valley degree of freedom. In this paper, we first review the recent progress on valley polarization in various TMD-based systems, including magnetically doped TMDs,intrinsic TMDs with both inversion and time-reversal symmetry broken, and magnetic TMD heterostructures. When topologically nontrivial bands are empowered into valley-polarized systems, valley-polarized topological states, namely valleypolarized quantum anomalous Hall effect can be realized. Therefore, we have also reviewed the theoretical proposals for realizing valley-polarized topological states in 2D honeycomb lattices. Our paper can help readers quickly grasp the latest research developments in this field.

Improved Variable Forgetting Factor Proportionate RLS Algorithm with Sparse Penalty and Fast Implementation Using DCD Iterations

The proportionate recursive least squares(PRLS)algorithm has shown faster convergence and better performance than both proportionate updating(PU)mechanism based least mean squares(LMS)algorithms and RLS algorithms with a sparse regularization term.In this paper,we propose a variable forgetting factor(VFF)PRLS algorithm with a sparse penalty,e.g.,l_(1)-norm,for sparse identification.To reduce the computation complexity of the proposed algorithm,a fast implementation method based on dichotomous coordinate descent(DCD)algorithm is also derived.Simulation results indicate superior performance of the proposed algorithm.

Global dust density in two-dimensional complex plasma

The driven-dissipative Langevin dynamics simulation is used to produce a two-dimensional(2D) dense cloud, which is composed of charged dust particles trapped in a quadratic potential. A 2D mesh grid is built to analyze the center-to-wall dust density. It is found that the local dust density in the outer region relative to that of the inner region is more nonuniform,being consistent with the feature of quadratic potential. The dependences of the global dust density on equilibrium temperature, particle size, confinement strength, and confinement shape are investigated. It is found that the particle size, the confinement strength, and the confinement shape strongly affect the global dust density, while the equilibrium temperature plays a minor effect on it. In the direction where there is a stronger confinement, the dust density gradient is bigger.

Boosting MA-based two-dimensional Ruddlesden-Popper perovskite solar cells by incorporating a binary spacer

Two-dimensional Ruddlesden-Popper(2DRP)perovskite exhibits excellent stability in perovskite solar cells(PSCs)due to introducing hydrophobic long-chain organic spacers.However,the poor charge transporting property of bulky organic cation spacers limits the performance of 2DRP PSCs.Inspired by the Asite cation alloying strategy in 3D perovskites,2DRP perovskites with a binary spacer can promote charge transporting compared to the unary spacer counterparts.Herein,the superior MA-based 2DRP perovskite films with a binary spacer,including 3-guanidinopropanoic acid(GPA)and 4-fluorophenethylamine(FPEA)are realized.These films(GPA_(0.85)FPEA_(0.15))_(2)MA_(4)Pb_5I_(16)show good morphology,large grain size,decreased trap state density,and preferential orientation of the as-prepared film.Accordingly,the present 2DRP-based PSC with the binary spacer achieves a remarkable efficiency of 18.37%with a V_(OC)of1.15 V,a J_(SC)of 20.13 mA cm^(-2),and an FF of 79.23%.To our knowledge,the PCE value should be the highest for binary spacer MA-based 2DRP(n≤5)PSCs to date.Importantly,owing to the hydrophobic fluorine group of FPEA and the enhanced interlayer interaction by FPEA,the unencapsulated 2DRP PSCs based on binary spacers exhibit much excellent humidity stability and thermal stability than the unary spacer counterparts.

Anomalous valley Hall effect in two-dimensional valleytronic materials

The anomalous valley Hall effect(AVHE)can be used to explore and utilize valley degrees of freedom in materials,which has potential applications in fields such as information storage,quantum computing and optoelectronics.AVHE exists in two-dimensional(2D)materials possessing valley polarization(VP),and such 2D materials usually belong to the hexagonal honeycomb lattice.Therefore,it is necessary to achieve valleytronic materials with VP that are more readily to be synthesized and applicated experimentally.In this topical review,we introduce recent developments on realizing VP as well as AVHE through different methods,i.e.,doping transition metal atoms,building ferrovalley heterostructures and searching for ferrovalley materials.Moreover,2D ferrovalley systems under external modulation are also discussed.2D valleytronic materials with AVHE demonstrate excellent performance and potential applications,which offer the possibility of realizing novel low-energy-consuming devices,facilitating further development of device technology,realizing miniaturization and enhancing functionality of them.

Low-frequency hybridized excess vibrations of two-dimensional glasses

One hallmark of glasses is the existence of excess vibrational modes at low frequenciesωbeyond Debye’s prediction.Numerous studies suggest that understanding low-frequency excess vibrations could help gain insight into the anomalous mechanical and thermodynamic properties of glasses.However,there is still intensive debate as to the frequency dependence of the population of low-frequency excess vibrations.In particular,excess modes could hybridize with phonon-like modes and the density of hybridized excess modes has been reported to follow D_(exc)(ω)~ω^(2)in 2D glasses with an inverse power law potential.Yet,the universality of the quadratic scaling remains unknown,since recent work suggested that interaction potentials could influence the scaling of the vibrational spectrum.Here,we extend the universality of the quadratic scaling for hybridized excess modes in 2D to glasses with potentials ranging from the purely repulsive soft-core interaction to the hard-core one with both repulsion and attraction as well as to glasses with significant differences in density or interparticle repulsion.Moreover,we observe that the number of hybridized excess modes exhibits a decrease in glasses with higher density or steeper interparticle repulsion,which is accompanied by a suppression of the strength of the sound attenuation.Our results indicate that the density bears some resemblance to the repulsive steepness of the interaction in influencing low-frequency properties.

Emerging two-dimensional Mo-based materials for rechargeable metal-ion batteries:Advances and perspectives

With the rapid development of rechargeable metal-ion batteries(MIBs)with safety,stability and high energy density,significant efforts have been devoted to exploring high-performance electrode materials.In recent years,two-dimensional(2D)molybdenum-based(Mo-based)materials have drawn considerable attention due to their exceptional characteristics,including low cost,unique crystal structure,high theoretical capacity and controllable chemical compositions.However,like other transition metal compounds,Mo-based materials are facing thorny challenges to overcome,such as slow electron/ion transfer kinetics and substantial volume changes during the charge and discharge processes.In this review,we summarize the recent progress in developing emerging 2D Mo-based electrode materials for MIBs,encompassing oxides,sulfides,selenides,carbides.After introducing the crystal structure and common synthesis methods,this review sheds light on the charge storage mechanism of several 2D Mo-based materials by various advanced characterization techniques.The latest achievements in utilizing 2D Mo-based materials as electrode materials for various MIBs(including lithium-ion batteries(LIBs),sodium-ion batteries(SIBs)and zinc-ion batteries(ZIBs))are discussed in detail.Afterwards,the modulation strategies for enhancing the electrochemical performance of 2D Mo-based materials are highlighted,focusing on heteroatom doping,vacancies creation,composite coupling engineering and nanostructure design.Finally,we present the existing challenges and future research directions for 2D Mo-based materials to realize high-performance energy storage systems.

Image segmentation of exfoliated two-dimensional materials by generative adversarial network-based data augmentation

Mechanically cleaved two-dimensional materials are random in size and thickness.Recognizing atomically thin flakes by human experts is inefficient and unsuitable for scalable production.Deep learning algorithms have been adopted as an alternative,nevertheless a major challenge is a lack of sufficient actual training images.Here we report the generation of synthetic two-dimensional materials images using StyleGAN3 to complement the dataset.DeepLabv3Plus network is trained with the synthetic images which reduces overfitting and improves recognition accuracy to over 90%.A semi-supervisory technique for labeling images is introduced to reduce manual efforts.The sharper edges recognized by this method facilitate material stacking with precise edge alignment,which benefits exploring novel properties of layered-material devices that crucially depend on the interlayer twist-angle.This feasible and efficient method allows for the rapid and high-quality manufacturing of atomically thin materials and devices.

SRMD:Sparse Random Mode Decomposition

Signal decomposition and multiscale signal analysis provide many useful tools for timefrequency analysis.We proposed a random feature method for analyzing time-series data by constructing a sparse approximation to the spectrogram.The randomization is both in the time window locations and the frequency sampling,which lowers the overall sampling and computational cost.The sparsification of the spectrogram leads to a sharp separation between time-frequency clusters which makes it easier to identify intrinsic modes,and thus leads to a new data-driven mode decomposition.The applications include signal representation,outlier removal,and mode decomposition.On benchmark tests,we show that our approach outperforms other state-of-the-art decomposition methods.