检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Qin, Zhaoming Dong, Nanqing Liu, Di Cao, Junwei Automatic Control Laboratory EPFL Lausanne1015 Switzerland Department of Computer Science University of Oxford OxfordOX1 3QD United Kingdom Department of Electrical Engineering Tsinghua University Beijing100084 China Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

As a data-driven approach, multi-agent reinforcement learning (MARL) has made remarkable advances in solving cooperative residential load scheduling problems. In a simplified scenario, centralized training, the most common paradigm for MARL, precludes large-scale deployment of MARL in communication-constrained cloud-edge environments. As a remedy, distributed training shows unparalleled advantages in real-life application scenarios compared with its centralized counterpart. However, distributed training still faces the challenge of system scalability, e.g., the high communication burden during coordinating individual agents, and needs to comply with data governance in terms of privacy. In this work, we propose a novel MARL solution to address these two practical issues. The proposed solution is based on actor-critic methods, where the global critic is a learned function of individual critics computed solely based on local observations of residences. In this way, the private data of residences is preserved completely and the communication cost is reduced significantly. The simulation experiments demonstrate that the proposed framework can achieve comparable performance to the state-of-the-art actor-critic framework without data governance and communication constraints. © 2023, The Authors. All rights reserved.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

computer vision for road imaging and pothole detection:a state-of-the-art review of systems and algorithms

引用

Transportation Safety and Environment 2022年第4期4卷 3-18页

作者： Nachuan Ma Jiahe Fan Wenshuo Wang JinWu Yu Jiang Lihua Xie Rui Fan Department of Control Science and Engineering Frontiers Science Center for Intelligent Autonomous Systemsand State Key Laboratory of Intelligent Autonomous SystemsTongji UniversityShanghai 201804P.R.China Department of Civil Engineering McGill UniversityMontréalQC H3A 0C3Canada Department of Electronics and Computer Engineering the Hong Kong University of Science and TechnologyHong Kong SAR 999077P.R.China CTO Office ClearMotion Inc.BillericaMA 01821USA School of Electrical and Electronic Engineering Nanyang Technological University50 Nanyang Avenue639798Singapore

computer vision algorithms have been utilized for 3-D road imaging and pothole detection for over two ***,there is a lack of systematic survey articles on state-of-the-art(SoTA)computer vision techniques,especially deep learningmodels,developed to tackle these *** article first introduces the sensing systems employed for 2-D and 3-D road data acquisition,including camera(s),laser scanners and Microsoft *** then comprehensively reviews the SoTA computer vision algorithms,including(1)classical 2-D image processing,(2)3-D point cloud modelling and segmentation and(3)machine/deep learning,developed for road pothole *** article also discusses the existing challenges and future development trends of computer vision-based road pothole detection approaches:classical 2-D image processing-based and 3-D point cloud modelling and segmentation-based approaches have already become history;and convolutional neural networks(CNNs)have demonstrated compelling road pothole detection results and are promising to break the bottleneck with future advances in self/un-supervised learning for multi-modal semantic *** believe that this survey can serve as practical guidance for developing the next-generation road condition assessment systems.

关键词： computer vision road imaging pothole detection deep learning image processing point cloud modelling convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

DSTFormer: 3D Human Pose Estimation with a Dual-scale Spatial and Temporal Transformer Network

DSTFormer: 3D Human Pose Estimation with a Dual-scale Spatia...

引用

International Conference on Advanced Robotics and Mechatronics (ICARM)

作者： Shaokun Zhang Xinde Li Chuanfei Hu Jianping Xu Huaping Liu School of Automation Southeast University Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China Nanjing Center for Applied Mathematics Nanjing China Southeast University Shenzhen Research Institute Shenzhen China Science and Technology on Information Systems Engineering Laboratory Nanjing China Department of Computer Science and Technology Tsinghua University Beijing China

ISBN: (数字)9798350385724

ISBN: (纸本)9798350385731

Recent transformer-based methods for estimating 3D human pose have gained widespread attention, achieving state-of-the-art results. Previous methods have primarily focused on capturing motion patterns of the human body at a single scale or cascading multiple scales, such as joints, bones, and body-parts. However, they are difficult to simultaneously capture spatial-temporal motion patterns of the human body at different scales due to the complex motion patterns. To address this issue, we propose Dual-scale Spatial and Temporal transFormer (DSTFormer), which can concurrently explore the spatial dependencies and temporal motion patterns of human joints and bones. Additionally, we introduce a Gcn-Spatial Transformer Block (GSTB), which introduces Graph Convolutional Networks (GCN) into transformer to enhance the exploitation of local relationships and global information between adjacent joints or bones. Extensive experiments are conducted on the Human3.6M benchmark dataset, and superior results are reported when comparing to other state-of-the-art methods. More remarkably, our model achieves to-date the best published performance, with P1 errors of 37.9 mm and 15.6 mm, respectively.

关键词： Solid modeling Three-dimensional displays Mechatronics Graph convolutional networks Computational modeling Pose estimation Transformers Bones Joints Robots

来源：评论

学校读者我要写书评

暂无评论

Diff9D: Diffusion-Based Domain-Generalized Category-Level 9-DoF Object Pose Estimation

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5520-5537页

作者： Jian Liu Wei Sun Hui Yang Pengchao Deng Chongpei Liu Nicu Sebe Hossein Rahmani Ajmal Mian National Engineering Research Center for Robot Visual Perception and Control Technology College of Electrical and Information Engineering School of Robotics and the State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body Hunan University Changsha China Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an China Department of Information Engineering and Computer Science University of Trento Trento Italy School of Computing and Communications Lancaster University Lancaster U.K. Department of Computer Science The University of Western Australia Crawley WA Australia

Nine-degrees-of-freedom (9-DoF) object pose and size estimation is crucial for enabling augmented reality and robotic manipulation. Category-level methods have received extensive research attention due to their potential for generalization to intra-class unknown objects. However, these methods require manual collection and labeling of large-scale real-world training data. To address this problem, we introduce a diffusion-based paradigm for domain-generalized category-level 9-DoF object pose estimation. Our motivation is to leverage the latent generalization ability of the diffusion model to address the domain generalization challenge in object pose estimation. This entails training the model exclusively on rendered synthetic data to achieve generalization to real-world scenes. We propose an effective diffusion model to redefine 9-DoF object pose estimation from a generative perspective. Our model does not require any 3D shape priors during training or inference. By employing the Denoising Diffusion Implicit Model, we demonstrate that the reverse diffusion process can be executed in as few as 3 steps, achieving near real-time performance. Finally, we design a robotic grasping system comprising both hardware and software components. Through comprehensive experiments on two benchmark datasets and the real-world robotic system, we show that our method achieves state-of-the-art domain generalization performance.

关键词： Shape Solid modeling Three-dimensional displays Pose estimation Training Robots Grasping Diffusion models Real-time systems Point cloud compression

来源：评论

学校读者我要写书评

暂无评论

Named Entity Recognition Based on Anchor Span for Manufacturing Text Knowledge Extraction

SSRN

引用

SSRN 2024年

作者： Li, Yahui Sun, Qi Zhou, Chunjie Liu, Lu Tian, Yu-Chu School of Artificial Intelligence and Automation Huazhong University of Science and Technology Hubei Wuhan430074 China Key Laboratory of Image Processing and Intelligent Control Ministry of Education Huazhong University of Science and Technology Hubei Wuhan430074 China School of Cyber Science and Engineering Huazhong University of Science and Technology Hubei Wuhan430074 China School of Computer Science Queensland University of Technology BrisbaneQLD4001 Australia

Intelligent industrial manufacturing heavily relies on structured knowledge. Named Entity Recognition (NER), an essential technique for extracting structured knowledge from text, has garnered significant research interest. However, current NER approaches face difficulties in handling multiple levels of entity nesting in knowledge extraction, especially within the manufacturing context. To address this issue, we present an Anchor Span-based NER (ASNER) approach for accurately and efficiently extracting manufacturing text knowledge. This method combines token and span classification. Initially, the head and tail features of the entity are extracted, with the corresponding anchor span of potential entities generated based on a boundary match neural network to localize the entities. Subsequently, token spatial contextual features are extracted using biaffine attention and convolutional neural networks, and classification categories are assigned to the anchor span-filtered entity features. Experimental studies demonstrate the effectiveness of the proposed ASNER approach. © 2024, The Authors. All rights reserved.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

Mitigating Low-Frequency Bias: Feature Recalibration and Frequency Attention Regularization for Adversarial Robustness

arXiv

引用

arXiv 2024年

作者： Zhang, Kejia Weng, Juanjuan Cai, Yuanzheng Luo, Zhiming Li, Shaozi Department of Artificial Intelligence Xiamen University Xiamen361005 China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control College of Computer and Control Engineering Minjiang University Fuzhou350108 China College of Information Science and Technology Jinan University Guangzhou510632 China

Ensuring the robustness of deep neural networks against adversarial attacks remains a fundamental challenge in computer vision. While adversarial training (AT) has emerged as a promising defense strategy, our analysis reveals a critical limitation: AT-trained models exhibit a bias toward low-frequency features while neglecting high-frequency components. This bias is particularly concerning as each frequency component carries distinct and crucial information: low-frequency features encode fundamental structural patterns, while high-frequency features capture intricate details and textures. To address this limitation, we propose High-Frequency Feature Disentanglement and Recalibration (HFDR), a novel module that strategically separates and recalibrates frequency-specific features to capture latent semantic cues. We further introduce frequency attention regularization to harmonize feature extraction across the frequency spectrum and mitigate the inherent low-frequency bias of AT. Extensive experiments demonstrate our method’s superior performance against white-box attacks and transfer attacks, while exhibiting strong generalization capabilities across diverse scenarios. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Fuzzy Sandwich control of Some Nonlinear Systems with a Number of Stochastically Occurring Impulses 13

Fuzzy Sandwich Control of Some Nonlinear Systems with a Numb...

引用

13th International Conference on Information science and technology, ICIST 2023

作者： Onasanya, Babatunde Oluwaseun Dai, Xiangguang Feng, Yuming Zhang, Wei School of Computer Science and Engineering Chongqing Three Gorges University Chongqing Wanzhou404100 China Department of Mathematics University of Ibadan Ibadan200284 Nigeria Chongqing Engineering Research Center of Internet of Things and Intelligent Control Technology Chongqing Three Gorges University Chongqing Wanzhou404100 China Three Gorges Institute of Artificial Intelligence Chongqing Three Gorges University Chongqing Wanzhou404100 China Key Laboratory of Intelligent Information Processing and Control Chongqing Three Gorges University Chongqing Wanzhou404100 China

ISBN: (纸本)9798350313925

In this paper, rather than have a fixed constant matrix in the sandwich control of the nonlinear system, a fuzzy control matrix was used. In addition, simple inequalities that do not require additional computational time and skills in computer programming were used rather than the usual LMI. The result of this method is that the period T required in the control process is shorter and the control time is also reached much faster or reduced compared to the existing sandwich methods that used constant matrices. In other words, the control time of this method is shorter than the control time achieved when using constant matrices. Besides, this method also can be used to control more nonlinear chaotic systems such as Lorenz's, Rössler's and Chen's systems. Numerical simulations were used to demonstrate the effectiveness of this method. © 2023 IEEE.

关键词： Nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

Adversarial Masking Contrastive Learning for vein recognition

arXiv

引用

arXiv 2024年

作者： Qin, Huafeng Wu, Yiquan El-Yacoubi, Mounim A. Wang, Jun Yang, Guangxiang Chongqing Engineering Laboratory of intelligent perception and block-chain School of Computer Science and Information Engineering Chongqing Technology and Business University Chongqing400067 China SAMOVAR Telecom SudParis CNRS University Paris-Saclay Paris91011 France School of Information and Control Engineering China University of Mining and Technology Xuzhou China

Vein recognition has received increasing attention due to its high security and privacy. Recently, deep neural networks such as Convolutional neural networks (CNN) and Transformers have been introduced for vein recognition and achieved state-of-the-art performance. Despite the recent advances, however, existing solutions for finger-vein feature extraction are still not optimal due to scarce training image samples. To overcome this problem, in this paper, we propose an adversarial masking contrastive learning (AMCL) approach, that generates challenging samples to train a more robust contrastive learning model for the downstream palm-vein recognition task, by alternatively optimizing the encoder in the contrastive learning model and a set of latent variables. First, a huge number of masks are generated to train a robust generative adversarial network (GAN). The trained generator transforms a latent variable from the latent variable space into a mask space. Then, we combine the trained generator with a contrastive learning model to obtain our AMCL, where the generator produces challenging masking images to increase the contrastive loss and the contrastive learning model is trained based on the harder images to learn a more robust feature representation. After training, the trained encoder in the contrastive learning model is combined with a classification layer to build a classifier, which is further fine-tuned on labeled training data for vein recognition. The experimental results on three databases demonstrate that our approach outperforms existing contrastive learning approaches in terms of improving identification accuracy of vein classifiers and achieves state-of-the-art recognition results. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Safe reinforcement learning task planning with uncertain duration and resource consumption in limited daytime for lunar rovers

Safe reinforcement learning task planning with uncertain dur...

引用

2024 IAF Space Operations Symposium at the 75th International Astronautical Congress, IAC 2024

作者： Lu, Siyao Xu, Rui Gao, Ai Li, Zhaoyu Liu, Jiamou Zhang, Libo Zhao, Zhijun Zhu, Shengying Li, Yuqiong School of Aerospace Engineering Beijing Institute of Technology 5th Zhongguancun South Street Haidian Distinct Beijing Beijing100081 China Key Laboratory of Autonomous Navigation and Control for Deep Space Exploration Beijing Institute of Technology Ministry of Industry and Information Technology Beijing100081 China School of Computer Science University of Auckland 34 Princes Street Auckland1010 New Zealand Beijing Institute of Spacecraft Engineering Beijing Beijing100094 China Institute of Mechanics Chinese Academy of Sciences 15th North Fourth Ring West Road Haidian District Beijing100190 China

ISBN: (纸本)9798331312183

The International Lunar Research Station will be established near the south pole through advanced unmanned rovers at the beginning period. The south pole of the moon has short daytime, so the efficiency of remote control is inadequate. However, the duration and power resource usage of the lunar rover moving on the lunar surface remains uncertain because of different loading weight of collection and changes of terrain in moving. What’s more, a lunar rover needs to move back to the base before nighttime without sunlight to provide energy, while the whole time of working on the moon also needs optimization. We select to solve the planning problem with reinforcement learning (RL) due to its capability in tackling uncertainty and optimization. However, traditional reinforcement learning cannot guarantee safety with time uncertainty, resource uncertainty, and constraints due to the soft constraints in optimization. Therefore, we propose a new way through safe reinforcement learning of task planning and resource collection optimization among tasks with uncertain duration and resource collection. We consider a scenario of in-situ material utilization for the lunar base, where there are tasks of moving, charging, collecting, material delivering, and material receiving, all of which have uncertain duration in execution and every task must be done during the daytime except the charging. Resource collection is related to power consumption in moving so it will be decided according to the remaining power. We further propose an architecture on reinforcement learning to let rovers decide the next step instantaneously according to the expected task duration, the remaining time, and the remaining power. Maximizing the amount of material delivered is an optimization target in training while keeping the rovers safe to work only in the daytime without an empty battery. In our experiment, we intend that our way works well in the uncertainties, and it will lead the rover to finish tasks w

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Large-Scale Data Generation Using SWLBM on the Sunway TaihuLight Supercomputer and Subsequent Data Mining with Physics-Informed Neural Networks 12th

Large-Scale Data Generation Using SWLBM on the Sunway TaihuL...

引用

12th CCF Conference on BigData, BigData 2024

作者： Chu, Xuesen Guo, Wei Wu, Tianqi Cai, Shengze Yang, Guangwen China Ship Scientific Research Center Wuxi214082 China Taihu Lake Laboratory of Deep Sea Technology and Science Wuxi214082 China Department of Computer Science and Technology Tsinghua University Beijing100084 China College of Control Science and Engineering Zhejiang University Hangzhou310027 China National Supercomputing Center in Wuxi Wuxi214072 China

ISBN: (纸本)9789819610235

Generating large-scale datasets on supercomputers is a critical component of modern research, enabling sophisticated analysis and mining applications. In the context of SUBOFF model studies for advanced submarine development, we emphasize the significance of supercomputer-generated big data and its subsequent application. Utilizing the Sunway TaihuLight supercomputer and the SWLBM software, we employ the lattice Boltzmann method (LBM) to generate a comprehensive dataset of flows over a SUBOFF model. This dataset is then used to train a physics-informed neural network (PINN) model, which aims to reconstruct flow fields from sparse velocity measurements. The PINN is designed to perform super-resolution velocity reconstruction and concurrent pressure field inference. Our results show that the reconstructed flow fields, including pressure, are in good agreement with the full-resolution LBM references, highlighting the promise of this approach for complex flow motion reconstruction. This research demonstrates the potential of supercomputer-generated big data in enhancing simulations and experiments in the study of SUBOFF and beyond. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Computational fluid dynamics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：