检索结果-内蒙古大学图书馆

41st International Conference on Machine Learning, ICML 2024

作者： Wang, Xin Lyu, Shengfei Yang, Lishan Zhan, Yibing Chen, Huanhuan School of Computer Science and Technology University of Science and Technology of China China Nanyang Technological University Singapore JD Explore Academy

Efficiently and effectively estimating counterfactuals over time is crucial for optimizing treatment strategies. We present the Adversarial Counterfactual Temporal Inference Network (ACTIN), a novel framework with dual modules to enhance counterfactual estimation. The balancing module employs a distribution-based adversarial method to learn balanced representations, extending beyond the limitations of current classification-based methods to mitigate confounding bias across various treatment types. The integrating module adopts a novel Temporal Integration Predicting (TIP) strategy, which has a wider receptive field of treatments and balanced representations from the beginning to the current time for a more profound level of analysis. TIP goes beyond the established direct Predicting (dP) strategy, which only relies on current treatments and representations, by empowering the integrating module to effectively capture long-range dependencies and temporal treatment interactions. ACTIN exceeds the confines of specific base models, and when implemented with simple base models, consistently delivers state-of-the-art performance and efficiency across both synthetic and real-world datasets. Copyright 2024 by the author(s)

关键词：

来源：评论

学校读者我要写书评

暂无评论

d-CONFOrMEr: deformable Sparse Transformer Augmented Convolution for Voxel-Based 3d Object detection 48

D-CONFORMER: Deformable Sparse Transformer Augmented Convolu...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Zhao, Xiao Su, Liuzhen Zhang, Xukun Yang, dingkang Sun, Mingyang Wang, Shunli Zhai, Peng Zhang, Lihua Fudan University Academy for Engineering and Technology Shanghai China Engineering Research Center of AI and Robotics Shanghai China

ISBN: (纸本)9781728163277

Although CNN-based and Transformer-based detectors have made impressive improvements in 3d object detection, these two network paradigms suffer from the interference of insufficient receptive field and local detail weakening, which significantly limits the feature extraction performance of the backbone. In this paper, we propose to fuse convolution and transformer, and simultaneously considering the different contributions of non-empty voxels at different positions in 3d space to object detection, it is not consistent with applying standard convolution and transformer directly on voxels. Specifically, we design a novel deformable sparse transformer to perform long-range information interaction on fine-grained local detail semantics aggregated by focal sparse convolution, termed d-Conformer. d-Conformer learns valuable voxels with position-wise in sparse space and can be applied to most voxel-based detectors as a backbone. Extensive experiments demonstrate that our method achieves satisfactory detection results and outperforms state-of-the-art 3d detection methods by a large margin. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Netrin-1 co-cross-linked hydrogel accelerates diabetic wound healing in situ by modulating macrophage heterogeneity and promoting angiogenesis

引用

Bioactive Materials 2024年第9期39卷 302-316页

作者： Futing Shu Hongchao Huang Shichu Xiao Zhaofan Xia Yongjun Zheng Department of Burn Surgery The First Affiliated Hospital of Naval Medical UniversityShanghai200433People’s Republic of China Research Unit of Key Techniques for Treatment of Burns and Combined Burns and Trauma Injury Chinese Academy of Medical SciencesShanghai200433People’s Republic of China

diabetic wounds,characterized by prolonged inflammation and impaired vascularization,are a serious complication of *** study aimed to design a gelatin methacrylate(GelMA)hydrogel for the sustained release of netrin-1 and evaluate its potential as a scaffold to promote diabetic wound *** results showed that netrin-1 was highly expressed during the inflammation and proliferation phases of normal wounds,whereas it synchronously exhibited aberrantly low expression in diabetic *** of netrin-1 inhibited normal wound healing,and the topical application of netrin-1 accelerated diabetic wound *** studies demonstrated that netrin-1 regulated macrophage heterogeneity via the A2br/STAT/PPArγsignaling pathway and promoted the function of endothelial cells,thus accelerating diabetic wound *** data suggest that netrin-1 is a potential therapeutic target for diabetic wounds.

关键词： Netrin-1 GelMA hydrogel diabetic wound healing Angiogenesis Macrophage heterogeneity

来源：评论

学校读者我要写书评

暂无评论

SA-P2T: Sparse Attention-based Pyramid Pooling Transformer Network for Object detection 24

SA-P2T: Sparse Attention-based Pyramid Pooling Transformer N...

引用

24th IEEE International Conference on Software Quality, reliability and Security Companion, QrS-C 2024

作者： Ma, Nan Xu, Genbao Han, Yiheng Sun, Beining Beijing University of Technology Faculty of Information Technology Beijing100124 China

ISBN: (纸本)9798350365658

recently, Transformer has demonstrated superior performance in object detection tasks by virtue of its powerful capability of modeling image global information. However, due to Transformer's expensive computational overhead and high memory occupation, the detection efficiency and performance are often affected when processing high-resolution or long sequence images. Later, some improvement works reduced the size of key and value in Transformer through pooling or convolution operations, which alleviated the above problems to a certain extent. Nonetheless, due to the smooth-ness of image data itself, the image patches in the smooth region do not contribute significantly to the final output of the model, resulting in data redundancy. Therefore, we propose a novel Sparse Attention-based Pyramid Pooling Transformer Network (SA-P2T) for object detection. Specifically, in SA-P2T, we introduce a sparse attention module, which measures the query sparsity via the Kullback-Leibler divergence, and then screen out the query that makes a large contribution to the self-attention as the new query. This module can filter out redundant information while retaining essential information, enabling the model to process image data more efficiently, and further reducing the computational complexity and memory space occupation of the model. Our experimental results on MS-COCO dataset illustrate that SA-P2T not only reduces computational complexity, but also enhances the accuracy and speed of detection, demonstrating the effectiveness of our proposed method. The code will be released at https://***/Genbao-Xu/SA-P2T. © 2024 IEEE.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

CAPTUrE: Efficient Attack Investigation Through Causal Alert Correlation Analysis 2nd

CAPTURE: Efficient Attack Investigation Through Causal Alert...

引用

2nd International Conference on Network Simulation and Evaluation, NSE 2023

作者： Chen, rongrong Ni, Xiaoya Xiao, Qianlong Li, Siying Hu, Minghao Xiao, Yanjun Qiu, Jing Guangzhou University Guangzhou510000 China NSFOCUS Technologies Group Co. Ltd Guangzhou510006 China Pengcheng Laboratory Shenzhen518000 China

ISBN: (纸本)9789819745180

Attack investigation is a vital approach for countering modern network attacks, which are often highly concealed and adversarial. From a proactive defense perspective, developing efficient and automated methods for attack investigation is of great significance as it enhances the ability to respond rapidly to complex network attacks. By representing system audit logs as provenance graphs that capture causal dependencies among attack events, and leveraging the strong correlation analysis and semantic capabilities of these graphs, we can effectively investigate complex and concealed network attacks, yielding substantial improvements compared to traditional methods. However, complex multi-stage network attacks can generate numerous alert points within the network. during the attack investigation process, analyzing this vast amount of alert data places a significant burden on human and computational resources, leading to what is known as "alert fatigue". To address this challenge, we propose CAPTUrE, a novel system designed to expedite attack investigation by employing causal alert correlation analysis in multi-alert scenarios. By optimizing the provenance graph through the search for alert-related subgraphs and connecting alert points using a reverse priority queue-based BFS search, CAPTUrE alleviates the investigative workload. Our experiments across 5 attack datasets demonstrate that CAPTUrE can effectively reduce the size of the investigation graph by 250 times. Furthermore, when compared to state-of-the-art methods, CAPTUrE reduces investigation time by 32.13%. Importantly, it accurately reconstructs complete attack scenarios. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

LIO-SAM for Vehicle Localization Using FGO Architecture 25

LIO-SAM for Vehicle Localization Using FGO Architecture

引用

25th International Conference on Electromagnetics in Advanced Applications, ICEAA 2024

作者： Yue, Long Lu, debiao Cai, Baigen Wang, Jian Liu, Jiang Jiang, Wei School of Automation and Intelligence Beijing Jiaotong University Beijing100044 China Beijing Engineering Center for EMC and GNSS Technologies Beijing100044 China Frontiers Science Center for Smart High-speed Railway System Beijing Jiaotong University Beijing100044 China

ISBN: (纸本)9798350360974

In intelligent transportation systems, accurate positioning is crucial for ensuring the safe and efficient operation of vehicles. This paper addresses the positioning challenges in satellite-constrained scenarios by proposing a solution based on Lidar-Inertial Odometry via Smoothing and Mapping (LIO-SAM). We employed the LIO-SAM algorithm as our core positioning mechanism and designed a multi-sensor fusion positioning architecture that integrates LidAr, Inertial Measurement Units (IMUs), and GNSS data to enhance positioning accuracy and robustness. By processing multisensor data obtained from on-site railway measurements, we validated and evaluated the feasibility and effectiveness of the algorithm. The experimental results indicate that this method maintains accurate and robust positioning even in satellite-constrained scenarios and achieves higher precision compared to traditional satellite positioning methods. This is significant for improving the safety and efficiency of vehicle operations. © 2024 IEEE.

关键词： Satellites

来源：评论

学校读者我要写书评

暂无评论

Influence of Pr-Al-Co alloy diffusion source on magnetic properties and microstructure of sintered Nd-Fe-B magnets processed by grain boundary diffusion

引用

Journal of rare Earths 2022年第12期40卷 1885-1893,I0003页

作者： Lei Jin Jinhao Zhu Guangfei ding Zhehuan Jin Bo Zheng Xuejing Cao Shuai Guo renjie Chen Aru Yan Key Laboratory of Magnetic Materials and Devices Ningbo Institute of Material Technologyand EngineeringChinese Academy of SciencesNingbo 315201China University of Chinese Academy of Sciences Beijing 100049China

In this study,the influence of the content of Al and Co in the diffusion source on the magnetic performance and microstructure of the diffused magnet was studied by grain boundary diffusion treatment with Pr_(70)Al_(30-x)Co_(x)(x=0 at%,10 at%,15 at%,20 at%,30 at%)*** the Co content in the diffusion source increases from 0 at%to 10 at%,the coercivity enhancement in the Pr_(70)Al_(20)Co_(10)diffused magnet is the highest,increased from 1.62 to 2.24 T,higher than 2.01 T of the Pr_(70)Al_(30)diffused *** further increase of Co content in the diffused source,the coercivity of the diffused magnet decreases gradually,the coercivity of Pr_(70)Al_(15)Co_(15),Pr_(70)Al_(20)Co_(10)and Pr_(70)Co_(30)diffused magnet is 2.15,1.99 and1.81 T,*** analysis shows that plenty of continuous grain boundary phases(CGBPs)can be formed in the Pr_(70)Al_(20)Co_(10)diffused magnet under the synergistic effect of Al and Co,which leads to the enhancement of magnetic isolation between more adjacent ***,the amount of CGBP in the diffused magnets gradually decreases with the further increase of Co content in the diffusion source.

关键词： Nd-Fe-B Coercivity Pr-AI-Co alloys Grain boundary diffusion rare earths

来源：评论

学校读者我要写书评

暂无评论

A review of research into the Standardization of High-Speed railway Text Operation and Maintenance data 6th

A Review of Research into the Standardization of High-Speed ...

引用

6th International Conference on Electrical Engineering and Information technologies for rail Transportation, EITrT 2023

作者： Tian, Shirun Yan, Fei School of Electronics and Information Engineering Beijing Jiaotong University Beijing100044 China Systems National Engineering Research Centre of Rail Transportation Operation and Control Beijing Jiaotong University Beijing100044 China

ISBN: (纸本)9789819993147

As the high-speed railway business in China develops rapidly, the safety and reliability requirements for high-speed railways are becoming higher and higher. For a high-speed railway system with a high degree of complexity, it is crucial to ensure the safe operation of the system from a risk management perspective by identifying potential accident hazards based on the collected risk source information and taking preventive measures in advance. This requires analysis of on-site incident and maintenance data to uncover correlations between sources of risk. Most of this data is recorded as textual data, in which there are many omissions, under-recording and multiple meanings of words. Although many relevant standard documents have been issued for different systems, there is a lack of a standardization management system for text data to achieve integrated management of risk source information for high-speed railway systems. This paper provides an overview of the key technologies that need to be applied in the standardization management of high-speed railway text data, and proposes a risk information management framework based on text information mining. © 2024, Beijing Paike Culture Commu. Co., Ltd.

关键词： Standardization

来源：评论

学校读者我要写书评

暂无评论

The research of the Control and Protection Adaptability for the Controllable Line Commutated Converter 20

The Research of the Control and Protection Adaptability for ...

引用

20th International Conference on AC and dC Power Transmission 2024, ACdC 2024

作者： Yin, Congqi Xu, Ying Li, Ming Wu, Fangjie Yang, Jun Zhang, Jing Liu, Hong Guo, Peiqian State Grid Economic and Technological Research Institute Co. Ltd. Changping Beijing China State Key Laboratory of Advanced Power Transmission Technology State Grid Smart Grid Research Institute Co. Ltd. Changping District Beijing China C-EPRI Electric Power Engineering Co. Ltd. Changping District Beijing China State Key Lab of Power Systems Dept. of Electrical Engineering Tsinghua University Haidian District Beijing China

ISBN: (纸本)9781837242672

The high capacity and long-distance transmission ability are the major advantages of the high voltage dC transmission system using Line Commutated Converter, which makes it one of the major transmission approaches in our country. However, the single-time or consecutive commutation failures during the AC system fault are the main drawbacks of the HVdC transmission using LCC valve, severely affecting the normal operation and power transmission. In this case, a new type of controllable line commutated converter (CLCC), which is embedded with the power electronics IGBT forcing commutation, is applied to avoid the commutation failure, therefore preventing the disturbance caused by the commutation failure like the traditional HVdC transmission system using LCC valve. This paper proposes the adaptive control and protection approaches according to the operation characteristics of the CLCC, and the optimizing approaches are also provided to improve the performance of the HVdC transmission system. This article provides reference for the world's first HVdC transmission system using CLCC. © The Institution of Engineering & Technology 2024.

关键词： HVdC power transmission

来源：评论

学校读者我要写书评

暂无评论

Surface Multiple Object Tracking: An Accurate HAT-YOLOv8-AdT Tracking Model

引用

IEEE Internet of Things Journal 2025年第11期12卷 18266-18278页

作者： Lin, Na Zhang, Lei Wu, Tianxiong Hawbani, Ammar Zhou, Huiyu Zhao, Liang Shenyang Aerospace University School of Computer Science Shenyang110136 China University of Leicester School of Computing and Mathematical Sciences LeicesterLE1 7RH United Kingdom

With the development of artificial intelligence technology, Autonomous aerial vehicles (AAV) have the ability to sense the environment. multiple object tracking (MOT) in AAV video is a very important vision task with a wide variety of applications. However, there are still many challenges in MOT in AAV video. First, the movement of the onboard camera in the three-dimensional (3-d) direction during the tracking process, as well as the unpredictable measurement noise characteristics of AAVs flying at high speeds, can lead to significant deviations in the prediction of the object’s position. Second, the applicability of the traditional detection algorithm decreases when the object is small and dense in the AAV viewpoint during detection. Finally, the traditional intersection over union (IoU) matching approach does not take into account the effects of the height and width of the box, and the matching results are inaccurate for the prediction and detection box. In order to address these challenges, we recommend the adaptive deepSort (AdT) algorithm to reduce the prediction bias due to camera movement and difficulty in predetermining measurement noise characteristics, the hybrid attention transformer-YOLOv8 (HAT-YOLOv8) algorithm to enhance the detection capability of tiny objects, and the IoU of height and width (HWIoU) matching algorithm, which improves the matching accuracy and thus the tracking accuracy. Experimental results show that our proposed solution outperforms the baseline solution. It outperforms the current mainstream StrongSort in MOTA, HOTA and IdF1 by 2.86%, 0.9%, and 9.36%. © 2014 IEEE.

关键词： Object tracking

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：