检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Zhuo Huang Miaoxi Zhu Xiaobo Xia Li Shen Jun Yu Chen Gong Bo Han Bo Du Tongliang Liu Sydney AI Centre The University of Sydney National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University JD Explore Academy Department of Automation University of Science and Technology of China Key Laboratory of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education School of Computer Science and Engineering Nanjing University of Science and Technology Department of Computer Science Hong Kong Baptist University

Robust generalization aims to tackle the most challenging data distributions which are rare in the training set and contain severe noises, i.e., photon-limited corruptions. Common solutions such as distributionally robust optimization (DRO) focus on the worst-case empirical risk to ensure low training error on the uncommon noisy distributions. However, due to the over-parameterized model being optimized on scarce worst-case data, DRO fails to produce a smooth loss landscape, thus struggling on generalizing well to the test set. Therefore, instead of focusing on the worst-case risk minimization, we propose SharpDRO by penalizing the sharpness of the worst-case distribution, which measures the loss changes around the neighbor of learning parameters. Through worst-case sharpness minimization, the proposed method successfully produces a flat loss curve on the corrupted distributions, thus achieving robust generalization. Moreover, by considering whether the distribution annotation is available, we apply SharpDRO to two problem settings and design a worst-case selection process for robust generalization. Theoretically, we show that SharpDRO has a great convergence guarantee. Experimentally, we simulate photon-limited corruptions using CIFAR10/100 and ImageNet30 datasets and show that SharpDRO exhibits a strong generalization ability against severe corruptions and exceeds well-known baseline methods with large performance gains.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic Pricing based Near-Optimal Resource Allocation for Elastic Edge Offloading

arXiv

引用

arXiv 2024年

作者： Xia, Yun Xue, Hai Zhang, Di Mumtaz, Shahid Xu, Xiaolong Rodrigues, Joel J.P.C. School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China School of Electrical and Information Engineering Zhengzhou University the Henan International Joint Laboratory of Intelligent Health Information System The National Telemedicine Center The National Engineering Laboratory for Internet Medical Systems and Applications Zhengzhou450001 China Department of Engineering Nottingham Trent University NottinghamNG1 4FQ United Kingdom The Department of Electronic Engineering Kyung Hee University Gyeonggi-do Yongin-si17104 Korea Republic of School of Software Nanjing University of Information Science and Technology Nanjing210044 China Amazonas State University AM Manaus Brazil

In mobile edge computing (MEC), task offloading can significantly reduce task execution latency and energy consumption of end user (EU). However, edge server (ES) resources are limited, necessitating efficient allocation to ensure the sustainable and healthy development for MEC systems. In this paper, we propose a dynamic pricing mechanism based near-optimal resource allocation for elastic edge offloading. First, we construct a resource pricing model and accordingly develop the utility functions for both EU and ES, the optimal pricing model parameters are derived by optimizing the utility functions. In the meantime, our theoretical analysis reveals that the EU’s utility function reaches a local maximum within the search range, but exhibits barely growth with increased resource allocation beyond this point. To this end, we further propose the Dynamic Inertia and Speed-Constrained particle swarm optimization (DISC-PSO) algorithm, which efficiently identifies the near-optimal resource allocation. Comprehensive simulation results validate the effectiveness of DISC-PSO, demonstrating that it significantly outperforms existing schemes by reducing the average number of iterations to reach a near-optimal solution by 92.11%, increasing the final user utility function value by 0.24%, and decreasing the variance of results by 95.45%. Copyright © 2024, The Authors. All rights reserved.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

SEVGGNET-LSTM: A FUSED DEEP LEARNING MODEL FOR ECG CLASSIFICATION

arXiv

引用

arXiv 2022年

作者： He, Tongyue Chen, Yiming Chen, Junxin Wang, Wei Zhou, Yicong College of Medicine and Biological Information Engineering Northeastern University Shenyang China School of Software Dalian University of Technology Dalian China School of Intelligent Systems Engineering Sun Yat-sen University Shenzhen China Department of Computer and Information Science University of Macau China

This paper presents a fused deep learning algorithm for ECG classification. It takes advantages of the combined convolutional and recurrent neural network for ECG classification, and the weight allocation capability of attention mechanism. The input ECG signals are firstly segmented and normalized, and then fed into the combined VGG and LSTM network for feature extraction and classification. An attention mechanism (SE block) is embedded into the core network for increasing the weight of important features. Two databases from different sources and devices are employed for performance validation, and the results well demonstrate the effectiveness and robustness of the proposed algorithm. Copyright © 2022, The Authors. All rights reserved.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

F2A2: flexible fully-decentralized approximate actor-critic for cooperative multi-agent reinforcement learning

The Journal of Machine Learning Research

引用

The Journal of Machine Learning Research 2023年第1期24卷 8447-8521页

作者： Wenhao Li Bo Jin Xiangfeng Wang Junchi Yan Hongyuan Zha School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of Software Engineering Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai China School of Computer Science and Technology Key Laboratory of Mathematics and Engineering Applications Ministry of Education East China Normal University Shanghai China Department of Computer Science and Engineering Key Laboratory of Artificial Intelligence Ministry of Education Shanghai Jiao Tong University Shanghai China

Traditional centralized multi-agent reinforcement learning (MARL) algorithms are sometimes unpractical in complicated applications due to non-interactivity between agents, the curse of dimensionality, and computation complexity. Hence, several decentralized MARL algorithms are motivated. However, existing decentralized methods only handle the fully cooperative setting where massive information needs to be transmitted in training. The block coordinate gradient descent scheme they used for successive independent actor and critic steps can simplify the calculation, but it causes serious bias. This paper proposes a exible fully decentralized actor-critic MARL framework, which can combine most of the actor-critic methods and handle large-scale general cooperative multi-agent settings. A primal-dual hybrid gradient descent type algorithm framework is designed to learn individual agents separately for decentralization. From the perspective of each agent, policy improvement and value evaluation are jointly optimized, which can stabilize multi-agent policy learning. Furthermore, the proposed framework can achieve scalability and stability for the large-scale environment. This framework also reduces information transmission by the parameter sharing mechanism and novel modeling-other-agents methods based on theory-of-mind and online supervised learning. Sufficient experiments in cooperative Multi-agent Particle Environment and StarCraft II show that the proposed decentralized MARL instantiation algorithms perform competitively against conventional centralized and decentralized methods.

关键词： cooperative MARL decentralized actor-critic primal-dual method

来源：评论

学校读者我要写书评

暂无评论

Low-Latency, Predictable, and High-Precision Networking for Cooperative Vehicle Infrastructure System

SSRN

引用

SSRN 2022年

作者： Lin, Chi Wu Yu, Chang Wang, Ning Institute of Intelligent System School of Software Dalian University of Technology Dalian China Department of Computer Science and Information Engineering Chung Hua University Hsinchu Taiwan Computer Science & Research Rowan University GlassboroNJ United States

Traditionally, infrastructural architects and road design have independently analyzed respective subsystems' needs in customary transportation. The two's inability to cooperate has culminated in possible safety issues, lousy road congestion, and unnecessary resource utilization. Moreover, challenges concerning sensor arrangement, data collecting, wireless communications, multi-source data fusion, analytical methodologies, and the development of decision-making processes might occur when cooperative vehicle infrastructure systems are used. In recent decades, Cooperative vehicle infrastructure systems(CVIS) have been suggested to assure intelligent transportation's effectiveness, effectiveness, and durability by integrating cars and infrastructures. A cooperative vehicle infrastructure system represents a novel business structure linked to businesses like autos, semiconductors, relevant information, and road infrastructure. Informatics, machine intelligence, telecommunications, data fusion, and intelligent monitoring are all included in this domain. Recent generations of networking technologies are pushing the boundaries of what became achievable utilizing cooperative vehicle infrastructure systems. Especially cooperative vehicle infrastructure systems, such instance, factory automation, Mixed Reality, and Touchscreen Network need ultra-low delay evaluated in single-digit milliseconds and cannot accept any fluctuation. Higher financial literacy requires an increased cloudification of application scenarios that require precise service level requirements for cooperative vehicle infrastructure systems. Almost all of those apps are mission-critical and therefore can risk perhaps a data packet being interrupted. Consequently, emerging applications need network infrastructure capable of delivering high-precision and predictable levels of service and accompanying assurances bordering on predictability. Cooperative vehicle infrastructure systems will require reassessing the f

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

RSAFormer: A method of polyp segmentation with region self-attention transformer

引用

Computers in Biology and Medicine 2024年 172卷 108268-108268页

作者： Yin, Xuehui Zeng, Jun Hou, Tianxiao Tang, Chao Gan, Chenquan Jain, Deepak Kumar García, Salvador School of Software Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China School of Cyber Security and Information Law Chongqing University of Posts and Telecommunications Chongqing400065 China Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education Dalian University of Technology Dalian116024 China Symbiosis Institute of Technology Symbiosis International University Pune412115 India Department of Computer Science and Artificial Intelligence Andalusian Research Institute in Data Science and Computational Intelligence University of Granada Granada18071 Spain

Colonoscopy has attached great importance to early screening and clinical diagnosis of colon cancer. It remains a challenging task to achieve fine segmentation of polyps. However, existing State-of-the-art models still have limited segmentation ability due to the lack of clear and highly similar boundaries between normal tissue and polyps. To deal with this problem, we propose a region self-attention enhancement network (RSAFormer) with a transformer encoder to capture more robust features. Different from other excellent methods, RSAFormer uniquely employs a dual decoder structure to generate various feature maps. Contrasting with traditional methods that typically employ a single decoder, it offers more flexibility and detail in feature extraction. RSAFormer also introduces a region self-attention enhancement module (RSA) to acquire more accurate feature information and foster a stronger interplay between low-level and high-level features. This module enhances uncertain areas to extract more precise boundary information, these areas being signified by regional context. Extensive experiments were conducted on five prevalent polyp datasets to demonstrate RSAFormer's proficiency. It achieves 92.2% and 83.5% mean Dice on Kvasir and ETIS, respectively, which outperformed most of the state-of-the-art models. © 2024 Elsevier Ltd

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Self-adaptive vision-language model for 3D segmentation of pulmonary artery and vein

arXiv

引用

arXiv 2025年

作者： Guo, Xiaotong Yang, Deqian Wang, Dan Zhao, Haochen Li, Yuan Sui, Zhilin Zhou, Tao Zhang, Lijun Meng, Yanda State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences China Department of Thoracic Surgery National Clinical Research Center for Cancer Cancer Hospital Shenzhen Hospital China School of Intelligent Science and Technology Hangzhou Institute for Advanced Study University of Chinese Academy of Sciences Hangzhou310024 China R&D Center Guangxi Huayi Artificial Intelligence Medical Technology Co. Ltd. China School of Computer Science and Engineering Beihang University Beijing China Department of Computer Science University of Exeter Exeter United Kingdom Department of Cardiovascular & Metabolic Medicine University of Liverpool Liverpool United Kingdom Guangzhou Jiayi Software Technology Co. Ltd. China

Accurate segmentation of pulmonary structures is crucial in clinical diagnosis, disease study, and treatment planning. Significant progress has been made in deep learning-based segmentation techniques, but most require much labeled data for training. Consequently, developing precise segmentation methods that demand fewer labeled datasets is paramount in medical image analysis. The emergence of pre-trained vision-language foundation models, such as CLIP, recently opened the door for universal computer vision tasks. Exploiting the generalization ability of these pre-trained foundation models on downstream tasks, such as segmentation, leads to unexpected performance with a relatively small amount of labeled data. However, exploring these models for pulmonary artery-vein segmentation is still limited. This paper proposes a novel framework called Language-guided self-adaptive Cross-Attention Fusion Framework. Our method adopts pre-trained CLIP as a strong feature extractor for generating the segmentation of 3D CT scans, while adaptively aggregating the cross-modality of text and image representations. We propose a s pecially designed adapter module to fine-tune pre-trained CLIP with a self-adaptive learning strategy to effectively fuse the two modalities of embeddings. We extensively validate our method on a local dataset, which is the largest pulmonary artery-vein CT dataset to date and consists of 718 labeled data in total. The experiments show that our method outperformed other state-of-the-art methods by a large margin. Our data and code will be made publicly available upon acceptance. Copyright © 2025, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Cognitive Disentanglement for Referring Multi-Object Tracking

arXiv

引用

arXiv 2025年

作者： Liang, Shaofeng Guan, Runwei Lian, Wangwang Liu, Daizong Sun, Xiaolou Wu, Dongming Yue, Yutao Ding, Weiping Xiong, Hui Qingdao China Guangzhou China Shandong Key Laboratory of Intelligent Oil & Gas Industrial Software Qingdao China Wangxuan Institute of Computer Technology Peking University Beijing China School of Automation Southeast University Nanjing China School of Computer Science Beijing Institute of Technology Beijing China School of Artificial Intelligence and Computer Science Nantong University Nantong China Guangzhou China Department of Electrical Engineering and Electronics University of Liverpool Liverpool United Kingdom

As a significant application of multi-source information fusion in intelligent transportation perception systems, Referring Multi-Object Tracking (RMOT) involves localizing and tracking specific objects in video sequences based on language references. However, existing RMOT approaches often treat language descriptions as holistic embeddings and struggle to effectively integrate the rich semantic information contained in language expressions with visual features. This limitation is especially apparent in complex scenes requiring comprehensive understanding of both static object attributes and spatial motion information. In this paper, we propose a Cognitive Disentanglement for Referring Multi-Object Tracking (CDRMT) framework that addresses these challenges. It adapts the "what" and "where" pathways from the human visual processing system to RMOT tasks. Specifically, our framework first establishes cross-modal connections while preserving modality-specific characteristics. It then disentangles language descriptions and hierarchically injects them into object queries, refining object understanding from coarse to fine-grained semantic levels. Finally, we reconstruct language representations based on visual features, ensuring that tracked objects faithfully reflect the referring expression. Extensive experiments on different benchmark datasets demonstrate that CDRMT achieves substantial improvements over state-of-the-art methods, with average gains of 6.0% in HOTA score on Refer-KITTI and 3.2% on Refer-KITTI-V2. Our approach advances the state-of-the-art in RMOT while simultaneously providing new insights into multi-source information fusion. Copyright © 2025, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

M3Care: Learning with Missing Modalities in Multimodal Healthcare Data

arXiv

引用

arXiv 2022年

作者： Zhang, Chaohe Chu, Xu Ma, Liantao Zhu, Yinghao Wang, Yasha Wang, Jiangtao Zhao, Junfeng Key Lab of High Confidence Software Technologies Ministry of Education School of Computer Science Peking University Beijing China Department of Computer Science and Technology Tsinghua University Key Lab of High Confidence Software Technologies Ministry of Education Beijing China Key Lab of High Confidence Software Technologies Ministry of Education National Engineering Research Center of Software Engineering Peking University Beijing China Center for Intelligent Healthcare Coventry University Coventry United Kingdom

Multimodal electronic health record (EHR) data are widely used in clinical applications. Conventional methods usually assume that each sample (patient) is associated with the unified observed modalities, and all modalities are available for each sample. However, missing modality caused by various clinical and social reasons is a common issue in real-world clinical scenarios. Existing methods mostly rely on solving a generative model that learns a mapping from the latent space to the original input space, which is an unstable ill-posed inverse problem. To relieve the underdetermined system, we propose a model solving a direct problem, dubbed learning with Missing Modalities in Multimodal healthcare data (M3Care). M3Care is an end-to-end model compensating the missing information of the patients with missing modalities to perform clinical analysis. Instead of generating raw missing data, M3Care imputes the task-related information of the missing modalities in the latent space by the auxiliary information from each patient's similar neighbors, measured by a task-guided modality-adaptive similarity metric, and thence conducts the clinical tasks. The task-guided modality-adaptive similarity metric utilizes the uncensored modalities of the patient and the other patients who also have the same uncensored modalities to find similar patients. Experiments on real-world datasets show that M3Care outperforms the state-of-the-art baselines. Moreover, the findings discovered by M3Care are consistent with experts and medical knowledge, demonstrating the capability and the potential of providing useful insights and explanations. Copyright © 2022, The Authors. All rights reserved.

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

Attention-Based Knowledge-aware Multi-Interest intelligent Model for Sequential Recommendation

Attention-Based Knowledge-aware Multi-Interest Intelligent M...

引用

Symposia and Workshops on Ubiquitous, Autonomic and Trusted Computing, UIC-ATC

作者： Yang Li Qianmu Li Shunmei Meng Jun Hou School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China Intelligent Manufacturing Department Wuyi University Jiangmen China State Key Lab. for Novel Software Technology Nanjing University Nanjing China School of Social Science Nanjing Vocational University of Industry Technology Nanjing China

As a fundamental means of intelligent service, recommender systems have shown a trend of diversification. Sequential recommendation, a class of time-aware methods, aims to predict the user’s next click. Multi-interest representation is a hot topic in recommender systems, but the research on how to apply multi-interest into sequences is still in the minority. And many methods will introduce knowledge graph to improve the representation of items, but how to consider both knowledge graph and user’s multi-interest in sequential recommendation has become a temporary blank. In this paper, we propose an attentive sequential recommendation with knowledge-aware multi-interest module (AKAMI). Items are embedded from the sequence’s structure and user’s interests. Specifically, we represent items by stacking self-attention layers to mine the internal correlation of the sequence. Meanwhile, we segment the sequence via the common neighbors in the knowledge graph and cast the user’s multi-interest based on the aggregation of subsequences. Multi-interests are integrated into item representation by the gating network adaptively. Comprehensive experiments on two real-world datasets show that the proposed model performs better than various representative methods under different evaluation metrics. Additionally, we also visually explain the model through a case study.

关键词：

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：