检索结果-南通市图书馆

维普期刊数据库

ControlService: a containerized solution for control-algorithm-as-a-service in cloud control systems

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第08期67卷 172-189页

作者： Chenggang SHAN Runze GAO Zhen YANG Wei ZHANG Yuanqing XIA School of artificial intelligence Zaozhuang University School of Automation Beijing Institute of Technology

As an extension of networked control systems, cloud control systems(CCSs) have emerged as a new control paradigm to improve the service quality of emerging control missions, such as data-driven modeling and automated vehicles. Existing studies have used the workflow-based restructured method to optimize the computation-intensive algorithms in the CCSs. However, the challenges here are how to define and submit these algorithms' workflows as cloud services and execute these algorithms' workflows in a containerized manner. Based on these challenges, we propose a containerized solution for the control-algorithm-as-a-service(C3aS) in the CCSs, namely Control Service. It offers the control algorithm as a cloud workflow service and uses a customized workflow engine to realize the containerized execution. First, we employ a cloud workflow representation method to define a control algorithm into an abstract cloud workflow form. Afterward, we provide a cloud service representation of the abstract cloud workflow. Next, we design a workflow engine and submit the cloud service to this workflow engine to implement containerized execution of this cloud service in the CCSs. In the experiment, we discuss the cloud service form and containerized implementation of the subspace identification method. Experimental results show that the proposed Control Service has significant performance advantages in computational time, reduction percentage, and speedup ratio compared with the baseline method.

关键词： cloud control systems control-algorithm-as-a-service containerization workflow engine cloud workflow service subspace identification

同方期刊数据库

在线全文

学校读者我要写书评

暂无评论

Optics-driven drone

Science China(Information Sciences) 2024年第2期67卷 322-324页

作者： Xuelong LI Guan HUANG Zhigang WANG Bin ZHAO School of artificial intelligence OPtics and Electro Nics (iOPEN)Northwestern Polytechnical University Shanghai artificial intelligence Laboratory

In the field of vicinagearth security, low-altitude flying vehicles such as drones play an indispensable role in aerial remote sensing, intelligent search and rescue, etc. [1]. It can be foreseen that the autonomous and intelligent execution of complex tasks by drones is an inevitable trend for future development.

关键词：

Ensemble successor representations for task generalization in offline-to-online reinforcement learning

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) 2024年第7期67卷 240-255页

作者： Changhong WANG Xudong YU Chenjia BAI Qiaosheng ZHANG Zhen WANG Space Control and Inertial Technology research Center Harbin Institute of Technology Shanghai artificial intelligence Laboratory Shenzhen research institute of Northwestern Polytechnical University School of Cybersecurity Northwestern Polytechnical University

In reinforcement learning(RL), training a policy from scratch with online experiences can be inefficient because of the difficulties in exploration. Recently, offline RL provides a promising solution by giving an initialized offline policy, which can be refined through online interactions. However, existing approaches primarily perform offline and online learning in the same task, without considering the task generalization problem in offline-to-online adaptation. In real-world applications, it is common that we only have an offline dataset from a specific task while aiming for fast online-adaptation for several tasks. To address this problem, our work builds upon the investigation of successor representations for task generalization in online RL and extends the framework to incorporate offline-to-online learning. We demonstrate that the conventional paradigm using successor features cannot effectively utilize offline data and improve the performance for the new task by online fine-tuning. To mitigate this, we introduce a novel methodology that leverages offline data to acquire an ensemble of successor representations and subsequently constructs ensemble Q functions. This approach enables robust representation learning from datasets with different coverage and facilitates fast adaption of Q functions towards new tasks during the online fine-tuning *** empirical evaluations provide compelling evidence showcasing the superior performance of our method in generalizing to diverse or even unseen tasks.

关键词： offline reinforcement learning online fine-tuning task generalization successor representations ensembles

同方期刊数据库

Key-Part Attention Retrieval for Robotic Object Recognition

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2024年第3期29卷 644-655页

作者： Jierui Liu Zhiqiang Cao Yingbo Tang State KeyLaboratory of Multimodal artificial intelligence Systems Institute of AutomationChinese Academy of SciencesBeijing100190China School of artificial intelligence University of Chinese Academy of SciencesBeijing 100049China

The ability to recognize novel objects with a few visual samples is critical in the robotic *** methods mainly concern the recognition of inter-category objects,however,the object recognition fromdifferent sub-classes within the same category remains challenging due to their similar *** thispaper,we propose a key-part attention retrieval solution to distinguish novel objects of different sub-classesaccording to a few samples without ***,an object encoder,including convolutional neuralnetwork with attention and key-part aggregation,is designed to generate object attention map and extract theobject-level embedding,where object attention map from the middle stage of the backbone is used to guide thekey-part ***,to overcome the non-differentiability drawback of key-part attention,the objectencoder is trained in a two-step scheme,and a more stable object-level embedding is *** this basis,the potential objects are located from a scene image by mining connected domains of the attention *** the embedding of each potential object and embeddings from support data,the recognition of thepotential objects is *** effectiveness of the proposed method is verified by experiments.

关键词： key-part attention retrieval robotic object recognition

A Bio-Inspired Integration Model of Basal Ganglia and Cerebellum for Motion Learning of a Musculoskeletal Robot

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

Journal of Systems Science & Complexity 2024年第1期37卷 82-113页

作者： ZHANG Jinhan CHEN Jiahao ZHONG Shanlin QIAO Hong State Key Laboratory of Multimodal artificial intelligence Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China School of artificial intelligence University of Chinese Academy of SciencesBeijing 100049China.

It is a significant research direction for highly complex musculoskeletal robots that how to develop the ability of motion learning and *** cooperations of multiple brain regions are crucial to improving motion *** by the neural mechanisms of structures,functions,and interconnections of basal ganglia and cerebellum,a biologically inspired integration model for motor learning of musculoskeletal robots is *** on the neural characteristics of the basal ganglia,the basal ganglia actor network,which mainly simulates the dorsal striatum,outputs motion commands,and the basal ganglia critic network,which simulates the ventral striatum,estimates actionstate *** network parameters are updated using the soft actor-critic *** on the sensorimotor prediction mechanism of the cerebellum,the cerebellum network evaluates the state feature extraction quality of the basal ganglia actor network and then updates the weights of its feature *** learning method is proven to converge to the optimal ***,drawing on the mechanism of dopaminergic dynamic regulation in the basal ganglia,the adaptive adjustment of target entropy and the dopaminergic experience replay are proposed to further improve the integration model,which contributes to the exploration-exploitation trade-off of motor *** bio-inspired integration model is validated on a musculoskeletal *** results indicate that this model can effectively control the musculoskeletal robot to accomplish the motion task from random starting locations to random target positions with high precision and robustness.

关键词： Basal ganglia and cerebellum bio-inspired integration model motion learning musculoskeletal robot reinforcement learning.

维普期刊数据库

Emergence of machine language: towards symbolic intelligence with neural networks

在线全文

学校读者我要写书评

暂无评论

National Science Review 2024年第4期11卷 27-30页

作者： Yuqi Wang Xu-Yao Zhang Cheng-Lin Liu Tieniu Tan Zhaoxiang Zhang State Key Laboratory of Multimodal artificial intelligence Systems Institute of AutomationChinese Academy of Sciences

Representation learning is a core issue in artificial intelligence (AI). Currently,there exists a disparity in the choice of representation between humans and machines. Humans rely on discrete language for communicati...

关键词：

Scene text recognition via dual character counting-aware visual and semantic modeling network

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) 2024年第3期67卷 313-314页

作者： Ke XIAO Anna ZHU Brian Kenji IWANA Cheng-Lin LIU School of Computer Science and artificial intelligence Wuhan University of Technology Human Interface Laboratory Kyushu University State Key Laboratory of Multimodal artificial intelligence Systems Institute of AutomationChinese Academy of Sciences School of artificial intelligence University of Chinese Academy of Sciences

Scene text recognition(STR) is drawing increasing attention nowadays due to its wide application in real life. Character counting information, as auxiliary information, has been shown to be effective in boosting text ... 详细信息

关键词： Decoding

同方期刊数据库

Non-Deterministic Liveness-Enforcing Supervisor Tolerant to Sensor-Reading Modification Attacks

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 240-248页

作者： Dan You Shouguang Wang IEEE the School of Information and Electronic Engineering Sussex Artificial Intelligence InstituteZhejiang Gongshang UniversityHangzhou 310018China

In this paper, we study the supervisory control problem of discrete event systems assuming that cyber-attacks might occur. In particular, we focus on the problem of liveness enforcement and consider a sensor-reading modification attack(SM-attack) that may disguise the occurrence of an event as that of another event by intruding sensor communication channels. To solve the problem, we introduce non-deterministic supervisors in the paper, which associate to every observed sequence a set of possible control actions offline and choose a control action from the set randomly online to control the system. Specifically, given a bounded Petri net(PN) as the reference formalism and an SMattack, an algorithm that synthesizes a liveness-enforcing nondeterministic supervisor tolerant to the SM-attack is proposed for the first time.

关键词： Cyber-attacks cyber-physical system(CPS) liveness non-deterministic supervisors Petri net(PN)

维普期刊数据库

Scattering-based hybrid network for facial attribute classification

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2024年第3期18卷 105-116页

作者： Na LIU Fan ZHANG Liang CHANG Fuqing DUAN School of artificial intelligence Beijing Normal UniversityBeijing 100875China

Face attribute classification(FAC)is a high-profile problem in biometric verification and face *** recent research has been devoted to extracting more delicate image attribute features and exploiting the inter-attribute correlations,significant challenges still *** scattering transform(WST)is a promising non-learned feature *** has been shown to yield more discriminative representations and outperforms the learned representations in certain *** to the image classification task,WST can enhance subtle image texture information and create local deformation *** paper designs a scattering-based hybrid block,to incorporate frequency-domain(WST)and image-domain features in a channel attention manner(Squeezeand-Excitation,SE),termed WS-SE *** with CNN,WS-SE achieves a more efficient FAC performance and compensates for the model sensitivity of the small-scale affine *** addition,to further exploit the relationships among the attribute labels,we propose a learning strategy from a causal *** cause attributes defined using the causalityrelated information can be utilized to infer the effect attributes with a high confidence *** analysis experiments demonstrate the effectiveness of our model,and our hybrid model obtains state-of-the-art results in two public datasets.

关键词： wavelet scattering transform causality-related learning facial attribute classification