检索结果-南通市图书馆

MetaCoorNet: an improved generated residual network for grasping pose estimation

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(information Sciences) 2025年第03期68卷 242-257页

作者： Hejia GAO Chuanfeng HE Junjie ZHAO Changyin SUN school of artificial intelligence Anhui University Engineering Research Center of Autonomous Unmanned System Technology Ministry of Education Anhui Provincial Key Laboratory of Security artificial intelligence Anhui University school of Automation Southeast University

Robotic grasping presents significant challenges due to variations in object properties, environmental complexities,and the demand for real-time operation. This study proposes the MetaCoorNet(MCN), which is a novel deep learning architecture specifically designed to address these challenges in robotic grasping pose estimation. By combining spatial and channel operators, the MetaCoor block is utilized to extract features efficiently. This architecture enhances feature selectivity by embedding location information into channel attention using a positional embedding technique within the coordinate attention mechanism. Consequently, the proposed MCN can focus on pertinent grasp-related regions. Furthermore, convolutional fusion blocks seamlessly integrate spatial and channel features, resulting in enhanced feature resolution and representation *** innovative design enables the proposed MCN to achieve state-of-the-art performance on the Cornell and Jacquard datasets,attaining accuracies of 98% and 91.2%, respectively. The effectiveness and robustness of MCN are further validated through real-world experiments conducted using a seven-degree-of-freedom Kinova manipulator.

关键词： generative ResNet meta light block coordinate attention feature resolution robot grasping

同方期刊数据库

Online Fault-Tolerant Tracking Control With Adaptive Critic for Nonaffine Nonlinear Systems

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2025年第1期12卷 215-227页

作者： Ding Wang Lingzhi Hu Xiaoli Li Junfei Qiao IEEE the school of information Science and Technology Beijing Key Laboratory of Computational Intelligence and Intelligent SystemBeijing Laboratory of Smart Environmental Protection and Beijing Institute of Artificial Intelligence Beijing University of Technology

In this paper, a fault-tolerant-based online critic learning algorithm is developed to solve the optimal tracking control issue for nonaffine nonlinear systems with actuator ***, a novel augmented plant is constructed by fusing the system state and the reference trajectory, which aims to transform the optimal fault-tolerant tracking control design with actuator faults into the optimal regulation problem of the conventional nonlinear error system. Subsequently, in order to ensure the normal execution of the online learning algorithm, a stability criterion condition is created to obtain an initial admissible tracking policy. Then, the constructed model neural network(NN) is pretrained to recognize the system dynamics and calculate trajectory control. The critic and action NNs are constructed to output the approximate cost function and approximate tracking control,respectively. The Hamilton-Jacobi-Bellman equation of the error system is solved online through the action-critic framework. In theoretical analysis, it is proved that all concerned signals are uniformly ultimately bounded according to the Lyapunov *** tracking control law can approach the optimal tracking control within a finite approximation error. Finally, two experimental examples are conducted to indicate the effectiveness and superiority of the developed fault-tolerant tracking control scheme.

关键词： Adaptive critic control neural network (NN) nonaffine nonlinear systems online fault-tolerant tracking design uniform ultimate boundedness

Fractal autoencoder with redundancy regularization for unsupervised feature selection

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(information Sciences) 2025年第2期68卷 89-102页

作者： Meiting SUN Fangyu LI Honggui HAN school of information Science and Technology Beijing University of Technology Beijing Key Laboratory of Computational intelligence and Intelligent System Engineering Research Center of Digital Community Ministry of Education Beijing Institute of artificial intelligence

Feature selection is a crucial step in data preprocessing because feature selection reduces the dimensionality of data by eliminating irrelevant and redundant features. Since manual labeling is expensive, unsupervised feature selection has received increasing attention in recent years. However, existing unsupervised feature selection methods tend to prioritize selecting highly correlated features over exploring feature diversity. Thus, a regularized fractal autoencoder(RFAE) method is proposed to select informative features in an unsupervised way. Specifically, the fractal autoencoder network extends autoencoders to construct a correspondence neural network and a selection neural network. The correspondence neural network exploits interfeature correlations and the selection neural network selects the informative features. A redundancy regularization strategy consists of a redundancy elimination regularization term based on the dependency between features and a sparse regularization term based on the group lasso. The redundancy regularization strategy eliminates feature subset redundancy and enhances network generalization ability. Extensive experimental results on six publicly available datasets show that the proposed RFAE outperforms the compared methods regarding clustering accuracy and classification accuracy. Moreover, the proposed RFAE achieves acceptable computation efficiency.

关键词： unsupervised feature selection fractal autoencoder correspondence neural network selection neural network redundancy regularization strategy

同方期刊数据库

Flexible artificial vision computing system based on FeOx optomemristor for speech recognition

在线全文

学校读者我要写书评

暂无评论

Journal of Semiconductors 2025年第1期46卷 225-232页

作者： Jie Li Yue Xin Bai Sun Dengshun Gu Changrong Liao Xiaofang Hu Lidan Wang Shukai Duan Guangdong Zhou College of artificial intelligence Southwest UniversityChongqing 400715China Frontier Institute of Science and Technology X’ian Jiaotong UniversityXi’an 710049China school of electronic information and Electrical Engineering Chongqing University of Arts and SciencesChongqing 402160China

With the advancement of artificial intelligence,optic in-sensing reservoir computing based on emerging semiconductor devices is high desirable for real-time analog signal ***,we disclose a flexible optomemristor based on C_(27)H_(30)O_(15)/FeOx heterostructure that presents a highly sensitive to the light stimuli and artificial optic synaptic features such as short-and long-term plasticity(STP and LTP),enabling the developed optomemristor to implement complex analogy signal processing through building a real-physical dynamic-based in-sensing reservoir computing algorithm and yielding an accuracy of 94.88%for speech *** charge trapping and detrapping mediated by the optic active layer of C_(27)H_(30)O_(15) that is extracted from the lotus flower is response for the positive photoconductance memory in the prepared *** work provides a feasible organic−inorganic heterostructure as well as an optic in-sensing vision computing for an advanced optic computing system in future complex signal processing.

关键词： reservoir computing flexible optomemristor analogy signal processing optic computing

维普期刊数据库

GPS: graph contrastive learning via multi-scale augmented views from adversarial pooling

在线全文

学校读者我要写书评

暂无评论

Science China(information Sciences) 2025年第1期68卷 145-158页

作者： Wei JU Yiyang GU Zhengyang MAO Ziyue QIAO Yifang QIN Xiao LUO Hui XIONG Ming ZHANG school of Computer Science National Key Laboratory for Multimedia Information ProcessingPeking University artificial intelligence Thrust The Hong Kong University of Science and Technology Department of Computer Science University of California

Self-supervised graph representation learning has recently shown considerable promise in a range of fields, including bioinformatics and social networks. A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs, which train models by maximizing agreement between original graphs and their augmented views(i.e., positive views). Unfortunately, these methods usually involve pre-defined augmentation strategies based on the knowledge of human experts. Moreover, these strategies may fail to generate challenging positive views to provide sufficient supervision signals. In this paper, we present a novel approach named graph pooling contrast(GPS) to address these *** by the fact that graph pooling can adaptively coarsen the graph with the removal of redundancy, we rethink graph pooling and leverage it to automatically generate multi-scale positive views with varying emphasis on providing challenging positives and preserving semantics, i.e., strongly-augmented view and weakly-augmented view. Then, we incorporate both views into a joint contrastive learning framework with similarity learning and consistency learning, where our pooling module is adversarially trained with respect to the encoder for adversarial robustness. Experiments on twelve datasets on both graph classification and transfer learning tasks verify the superiority of the proposed method over its counterparts.

关键词： graph representation learning graph neural networks graph contrastive learning graph augmentations graph pooling

Text-Image Feature Fine-Grained Learning for Joint Multimodal Aspect-Based Sentiment Analysis

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2025年第1期82卷 279-305页

作者： Tianzhi Zhang Gang Zhou Shuang Zhang Shunhang Li Yepeng Sun Qiankun Pi Shuo Liu school of Data and Target Engineering Information Engineering UniversityZhengzhou450001China information Engineering Department Liaoning Provincial College of CommunicationsShenyang110122China school of Computer and artificial intelligence Zhengzhou UniversityZhengzhou450000China

Joint Multimodal Aspect-based Sentiment Analysis(JMASA)is a significant task in the research of multimodal fine-grained sentiment analysis,which combines two subtasks:Multimodal Aspect Term Extraction(MATE)and Multimodal Aspect-oriented Sentiment Classification(MASC).Currently,most existing models for JMASA only perform text and image feature encoding from a basic level,but often neglect the in-depth analysis of unimodal intrinsic features,which may lead to the low accuracy of aspect term extraction and the poor ability of sentiment prediction due to the insufficient learning of intra-modal *** this problem,we propose a Text-Image Feature Fine-grained Learning(TIFFL)model for ***,we construct an enhanced adjacency matrix of word dependencies and adopt graph convolutional network to learn the syntactic structure features for text,which addresses the context interference problem of identifying different aspect ***,the adjective-noun pairs extracted from image are introduced to enable the semantic representation of visual features more intuitive,which addresses the ambiguous semantic extraction problem during image feature ***,the model performance of aspect term extraction and sentiment polarity prediction can be further optimized and *** on two Twitter benchmark datasets demonstrate that TIFFL achieves competitive results for JMASA,MATE and MASC,thus validating the effectiveness of our proposed methods.

关键词： Multimodal sentiment analysis aspect-based sentiment analysis feature fine-grained learning graph convolutional network adjective-noun pairs

An Effective Intrusion Detection System Based on the FSA-BGRU Hybrid Model

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2025年第2期22卷 188-198页

作者： Deng Zaihui Li Zihang Guo Jianzhong Gan Guangming Kong Dejin school of electronic and Electrical Engineering Wuhan Textile UniversityWuhan 430200China school of Computer Science and artificial intelligence Wuhan Textile UniversityWuhan 430200China school of Physics and electronic information Engineering Hubei Engineering UniversityXiaogan 432000China

Intrusion detection systems play a vital role in cyberspace *** this study,a network intrusion detection method based on the feature selection algorithm(FSA)and a deep learning model is developed using a fusion of a recursive feature elimination(RFE)algorithm and a bidirectional gated recurrent unit(BGRU).Particularly,the RFE algorithm is employed to select features from high-dimensional data to reduce weak correlations between features and remove redundant features in the numerical feature ***,a neural network that combines the BGRU and multilayer perceptron(MLP)is adopted to extract deep intrusion behavior ***,a support vector machine(SVM)classifier is used to classify intrusion *** proposed model is verified by experiments on the NSL-KDD *** results indicate that the proposed model achieves a 90.25%accuracy and a 97.51%detection rate in binary classification and outperforms other machine learning and deep learning models in intrusion *** proposed method can provide new insight into network intrusion detection.

关键词： bidirectional GRU feature selection intrusion detection system multilayer perceptron recursive feature elimination support vector machine

维普期刊数据库

Smaller, faster, lower-power analog RRAM matrix computing circuits without performance compromise

在线全文

学校读者我要写书评

暂无评论

Science China(information Sciences) 2025年第2期68卷 293-306页

作者： Yubiao LUO Pushen ZUO Shiqing WANG Zhong SUN Ru HUANG school of Integrated Circuits Peking University Institute for artificial intelligence Peking University Beijing Advanced Innovation Center for Integrated Circuits

Recently, the analog matrix computing(AMC) concept has been proposed for fast, efficient matrix operations, by configuring global feedback loops with crosspoint resistive memory arrays and operational amplifiers(OAs). The implementation of a general real-valued matrix(containing both positive and negative elements) is enabled by using a set of analog inverters,which, however, is considered inefficient regarding circuit compactness, power consumption, and temporal response. Here, with the assistance of the conductance compensation(CC) strategy to take full advantage of the inherent differential inputs of OAs,new AMC circuits without analog inverters are designed. Such a design saves the area occupation and power dissipation of analog inverters, and thus turns to be smaller and lower-power. Simulation results reveal that the new circuit also shows a faster response towards the steady state, thanks to the reduction of poles in the circuit, which, again, is contributed by the elimination of analog inverters. Along with all of these benefits, extensive simulations demonstrate that the CC-AMC circuits do not compromise the computing performance in terms of relative error caused by various non-ideal factors in the circuit.

关键词： analog computing in-memory computing matrix resistive memory conductance compensation

同方期刊数据库

Carbon interconnected microsized Si film toward high energy room temperature solid-state lithium-ion batteries

在线全文

学校读者我要写书评

暂无评论

Chinese Chemical Letters 2025年第2期36卷 531-535页

作者： Xin Li Ling Zhang Yunyan Fan Shaojing Lin Yong Lin Yongsheng Ying Meijiao Hu Haiying Gao Xianri Xu Zhongbiao Xia Xinchuan Lin Junjie Lu Xiang Han school of information Engineering Fujian Business University College of artificial intelligence Yango University College of Materials Science and Engineering Co-Innovation Center of Efficient Processing and Utilization of Forest Resources Nanjing Forestry University

Solid-state batteries （SSBs） with high-capacity Si anodes have been regarded as one of the most promising candidates to meet the large scale energy storage and electrical vehicles due to its intrinsic safety and potential high energy ***,Si suffers from poor electrical conductivity and huge volume change and particles fracture during lithiaiotn and delithiation,which induces low practical energy *** addition,the SSBs are often operated at high temperature due to the poor physical contact and huge resistance between Si and solid-state electrolyte （SSE）.To improve the bulk electronic/ionic conductivity of Si and its interfacial compatibility with SSE,herein,a binder free and self-supporting Si/C film was *** monolithic carbon not only enhance the electric conductivity but also release huge stress during lithiation and *** addition,paired with the flexible and soft poly（vinylidene fluoride）-co-hexafluoropropylene （PVDF-HFP） and Li1.3Al0.3Ti1.7（PO4）3（LATP） solid-state electrolyte,a LiF-rich and electrochemical stable solid-electrolyte interphase （SEI） layer is in-situ *** fast bulk and interfacial ionic transportation as well as the mechanical integrity of MSi enable high performance SSBs at room *** a result,high specific capacity of 2137 m Ah/g with an initial Coulombic efficiency of 83.2%is obtained at a rate of 0.5 A/*** at a high rate of 3 A/g,the specific capacity is1793 m Ah/*** a rate of 1 A/g,the Si/C anode delivers a long cycling performance over 500 cycles while maintains a capacity of 1135 mAh/*** work provides a new strategy that combines charge transfer kinetics and interfacial chemistry design toward high energy density Si-based SSBs.

关键词： Carbon coating Interphase Lithium-ion battery Silicon anode Solid-state electrolyte