检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Song, Ziying Zhang, Guoxin Xie, Jun Liu, Lin Jia, Caiyan Xu, Shaoqing Wang, Zhepeng School of Computer and Information Technology Beijing Key Lab of Traffic Data Analysis and Mining Beijing Jiaotong University Beijing100044 China Hebei University of Science and Technology School of Information Science and Engineering Shijiazhuang050018 China Lenovo Research Beijing100085 China State Key Laboratory of Internet of Things for Smart City Department of Electrome chanical Engineering University of Macau 999078 China

LiDAR-camera fusion can enhance the performance of 3D object detection by utilizing complementary information between depth-aware LiDAR points and semantically rich images. Existing voxel-based methods face significant challenges when fusing sparse voxel features with dense image features in a one-to-one manner, resulting in the loss of the advantages of images, including semantic and continuity information, leading to sub-optimal detection performance, especially at long distances. In this paper, we present VoxelNextFusion, a multi-modal 3D object detection framework specifically designed for voxel-based methods, which effectively bridges the gap between sparse point clouds and dense images. In particular, we propose a voxel-based image pipeline that involves projecting point clouds onto images to obtain both pixel- and patch-level features. These features are then fused using a self-attention to obtain a combined representation. Moreover, to address the issue of background features present in patches, we propose a feature importance module that effectively distinguishes between foreground and background features, thus minimizing the impact of the background features. Extensive experiments were conducted on the widely used KITTI and nuScenes 3D object detection benchmarks. Notably, our VoxelNextFusion achieved around +3.20% in AP@0.7 improvement for car detection in hard level compared to the Voxel R-CNN baseline on the KITTI test dataset. Copyright © 2024, 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

CLAPSep: Leveraging Contrastive Pre-trained Model for Multi-Modal Query-Conditioned Target Sound Extraction

arXiv

引用

arXiv 2024年

作者： Ma, Hao Peng, Zhiyuan Li, Xu Shao, Mingjie Wu, Xixin Liu, Ju School of Information Science and Engineering Shandong University Qingdao China Department of Computer Science North Carolina State University NC United States ARC Lab Tencent PCG China Stanley Ho Big Data Decision Analytics Research Centre The Chinese University of Hong Kong Hong Kong

Universal sound separation (USS) aims to extract arbitrary types of sounds from real-world recordings. This can be achieved by language-queried target sound extraction (TSE), which typically consists of two components: a query network that converts user queries into conditional embeddings, and a separation network that extracts the target sound accordingly. Existing methods commonly train models from scratch. As a consequence, substantial data and computational resources are required to make the randomly initialized model comprehend sound events and perform separation accordingly. In this paper, we propose to integrate pre-trained models into TSE models to address the above issue. To be specific, we tailor and adapt the powerful contrastive language-audio pre-trained model (CLAP) for USS, denoted as CLAPSep. CLAPSep also accepts flexible user inputs, taking both positive and negative user prompts of uni- and/or multi-modalities for target sound extraction. These key features of CLAPSep can not only enhance the extraction performance but also improve the versatility of its application. We provide extensive experiments on 5 diverse datasets to demonstrate the superior performance and zero- and few-shot generalizability of our proposed CLAPSep with fast training convergence, surpassing previous methods by a significant margin. Full codes and some audio examples are released for reproduction and evaluation1 © 2024, CC BY.

关键词： Query languages

来源：评论

学校读者我要写书评

暂无评论

Fast Latent Factor Analysis via a Fuzzy PID-Incorporated Stochastic Gradient Descent Algorithm

arXiv

引用

arXiv 2023年

作者： Li, Jinli Yuan, Ye School of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing400065 China Chongqing Key Laboratory of Big Data and Intelligent Computing Chongqing Engineering Research Center of Big Data Application for Smart Cities Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences Chongqing400714 China College of Computer and Information Science Southwest University Chongqing400715 China

A high-dimensional and incomplete (HDI) matrix can describe the complex interactions among numerous nodes in various big data-related applications. A stochastic gradient descent (SGD)-based latent factor analysis (LFA) model is remarkably effective in extracting valuable information from an HDI matrix. However, such a model commonly encounters the problem of slow convergence because a standard SGD algorithm learns a latent factor relying on the stochastic gradient of current instance error only without considering past update information. To address this critical issue, this paper innovatively proposes a Fuzzy PID-incorporated SGD (FPS) algorithm with two-fold ideas: 1) rebuilding the instance learning error by considering the past update information in an efficient way following the principle of PID, and 2) implementing hyper-parameters and gain parameters adaptation following the fuzzy rules. With it, an FPS-incorporated LFA model is further achieved for fast processing an HDI matrix. Empirical studies on six HDI datasets demonstrate that the proposed FPS-incorporated LFA model significantly outperforms the state-of-the-art LFA models in terms of computational efficiency for predicting the missing data of an HDI matrix with competitive accuracy. Copyright © 2023, The Authors. All rights reserved.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Improving Fast Adversarial Training Paradigm: An Example Taxonomy Perspective

arXiv

引用

arXiv 2024年

作者： Gui, Jie Jiang, Chengze Dong, Minjing Tong, Kun Shi, Xinli Tang, Yuan Yan Tao, Dacheng The School of Cyber Science and Engineering Southeast University Nanjing210000 China Purple Mountain Laboratories Nanjing210000 China The Department of Computer Science City University of Hong Kong Hong Kong The Department of Computer and Information Science University of Macau 999078 China The College of Computing & Data Science Nanyang Technological University #32 Block N4 #02a-014 50 Nanyang Avenue Singapore639798 Singapore

While adversarial training is an effective defense method against adversarial attacks, it notably increases the training cost. To this end, fast adversarial training (FAT) is presented for efficient training and has become a hot research topic. However, FAT suffers from catastrophic overfitting, which leads to a performance drop compared with multi-step adversarial training. However, the cause of catastrophic overfitting remains unclear and lacks exploration. In this paper, we present an example taxonomy in FAT, which identifies that catastrophic overfitting is caused by the imbalance between the inner and outer optimization in FAT. Furthermore, we investigated the impact of varying degrees of training loss, revealing a correlation between training loss and catastrophic overfitting. Based on these observations, we redesign the loss function in FAT with the proposed dynamic label relaxation to concentrate the loss range and reduce the impact of misclassified examples. Meanwhile, we introduce batch momentum initialization to enhance the diversity to prevent catastrophic overfitting in an efficient manner. Furthermore, we also propose Catastrophic Overfitting aware Loss Adaptation (COLA), which employs a separate training strategy for examples based on their loss degree. Our proposed method, named example taxonomy aware FAT (ETA), establishes an improved paradigm for FAT. Experiment results demonstrate our ETA achieves state-of-the-art performance. Comprehensive experiments on four standard datasets demonstrate the competitiveness of our proposed method. 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

Medical Visual Question Answering with Symmetric Interaction Attention and Cross-Modal Gating

SSRN

引用

SSRN 2023年

作者： Chen, Zhi Zou, Beiji Dai, Yulan Zhu, Chengzhang Kong, Guilan Zhang, Wensheng School of Computer Science and Engineering Central South University Changsha410083 China National Institute of Health Data Science Peking University Beijing100871 China Instiute of Automation Chinese Academy of Sciences Beijing100190 China College of Literature and Journalism

The purpose of medical visual question answering (Med-VQA) is to provide accurate answers to clinical questions related to visual content of medical images. However, previous attempts neglect to take full advantage of the information interaction between medical images and clinical questions, which hinders the further progress of Med-VQA. The above issue requires the efforts to focus on critical information interaction within each modality and relevant information interaction between modalities. In this paper, we first design a symmetric interaction attention module to construct dense and deep intra- and inter-modal information interaction on medical images and clinical questions for the Med-VQA task. Specifically, the symmetric interaction attention module consists of multiple symmetric interaction attention blocks that contain two basic units, i.e., self-attention and interaction attention. Technically, self-attention is introduced for intra-modal information interaction, while interaction attention is constructed for inter-modal information interaction. Then, we develop a multi-modal fusion scheme that leverages the cross-modal gating to effectively fuse multi-modal information and avoid redundant information after sufficient intra- and inter-modal information interaction. Experimental results on the VQA-RAD dataset show that our method outperforms other state-of-the-art Med-VQA models, obtaining 81.2%, 65.4% and 74.9% accuracy for close-ended answers, open-ended answers, and overall test data, respectively. Furthermore, our approach shows great superiority in improving the accuracy of open-ended answers and improves 4.5% absolute accuracy compared to the baseline model. It is expected to be applied in computer-assisted medical diagnosis and help relieve the pressure of radiologists. © 2023, The Authors. All rights reserved.

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

WMFormer++: Nested Transformer for Visible Watermark Removal via Implict Joint Learning

arXiv

引用

arXiv 2023年

作者： Huo, Dongjian Zhang, Zehong Su, Hanjing Li, Guanbin Fang, Chaowei Wu, Qingyao School of Software and Engineering South China University of Technology China Key Laboratory of Big Data and Intelligent Robot Ministry of Education Peng Cheng Laboratory China Tencent Wechat Department School of Computer Science and Engineering Sun Yat-sen University China Xidian University China

Watermarking serves as a widely adopted approach to safeguard media copyright. In parallel, the research focus has extended to watermark removal techniques, offering an adversarial means to enhance watermark robustness and foster advancements in the watermarking field. Existing watermark removal methods mainly rely on UNet with task-specific decoder branches——one for watermark localization and the other for background image restoration. However, watermark localization and background restoration are not isolated tasks;precise watermark localization inherently implies regions necessitating restoration, and the background restoration process contributes to more accurate watermark localization. To holistically integrate information from both branches, we introduce an implicit joint learning paradigm. This empowers the network to autonomously navigate the flow of information between implicit branches through a gate mechanism. Furthermore, we employ cross-channel attention to facilitate local detail restoration and holistic structural comprehension, while harnessing nested structures to integrate multiscale information. Extensive experiments are conducted on various challenging benchmarks to validate the effectiveness of our proposed method. The results demonstrate our approach’s remarkable superiority, surpassing existing state-ofthe-art methods by a large margin. Copyright © 2023, The Authors. All rights reserved.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Prediction of Startup Performance Using Bidirectional Long Short-Term Memory with Parametric Switch Activation Function

Prediction of Startup Performance Using Bidirectional Long S...

引用

Ambient Intelligence, Knowledge Informatics and Industrial Electronics (AIKIIE), International Conference on

作者： Putta Srivani Gopu Sowjanya Bura Vijay Kumar Kassem AL-Attabi Ibraheem Hatem Mohammed Department of CSE (Cyber Security) Malla Reddy Engineering college for women Hyderabad India Department of CSE(Data Science) CMR Technical Campus Hyderabad India School of Computer Science and Artificial Intelligence SR University Warangal India The Islamic University Najaf Iraq Department of Computer Communications Engineering Al Rafidain University College Bagdad Iraq

Predicting start-up success has garnered increasing attention in recent years. Numerous predictive models and methodologies have been devised to ascertain the probability of a start-up's triumph contingent on diverse aspects. Large datasets have been analysed using machine learning techniques like logistic regression, decision trees, and neural networks to find patterns that can help predict startup success. Predicting the success of a start-up, however, is a difficult undertaking because the business environment is unpredictable and changeable. Efficient resource utilization is crucial for economic development and has a direct impact on reducing unemployment rates. In order to minimize resource wastage and mitigate the risk of failure, the research identified and examined specific factors that influence the success or failure of small-scale companies. In the current research, the Crunch Base dataset is utilized to predict the failure or success of start-ups by employing the Bidirectional Long Short-Term Memory (Bi-LSTM) with Parametric Switch Activation (PSA) function model. The Bi-LSTM unit, a type of Recurrent Neural Network (RNN), incorporates the Feed Forward Neural (FFN) Network for classification purposes. The suggested Bi-LSTM model achieved a higher accuracy rate of 72.26% compared to existing methods. The suggested Bi-LSTM model achieved a higher accuracy rate of 72.26% compared to existing methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Theoretical Analysis of Self-Supervised Learning for Vision Transformers

arXiv

引用

arXiv 2024年

作者： Huang, Yu Wen, Zixin Chi, Yuejie Liang, Yingbin UPenn United States CMU United States OSU United States Department of Statistics and Data Science Wharton School University of Pennsylvania United States Machine Learning Department Carnegie Mellon University United States Department of Electrical and Computer Engineering Carnegie Mellon University United States Department of Electrical and Computer Engineering The Ohio State University United States

Self-supervised learning (SSL) has become a foundational approach in computer vision, which is broadly categorized into reconstruction-based methods like masked autoencoders (MAE) and discriminative methods such as contrastive learning (CL). Recent empirical observations reveal that MAE and CL capture distinct representations. CL tends to focus on global patterns, while MAE adeptly captures both global and subtle local information simultaneously. Despite a flurry of recent empirical investigations to shed light on this difference, theoretical understanding remains limited, especially on the dominant architecture: vision transformers (ViTs). In this paper, to provide rigorous insights, we model the distribution of visual data by considering two types of spatial features: dominant global features and comparatively minuscule local features and study the impact of imbalance among these features. We analyze the training dynamics of one-layer softmax-based ViTs on both MAE and CL objectives using gradient descent. Our analysis shows that as the degree of feature imbalance varies, ViTs trained with the MAE objective effectively learn both global and local features to achieve near-optimal reconstruction, while the CL-trained ViTs favor predominantly global features, even under mild imbalance. These results provide a theoretical explanation for the different behaviors of MAE and CL observed in prior studies. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Energy-based Proportional Fairness in Cooperative Edge Computing

arXiv

引用

arXiv 2022年

作者： Vu, Thai T. Chu, Nam H. Phan, Khoa T. Hoang, Dinh Thai Nguyen, Diep N. Dutkiewicz, Eryk School of Engineering and Mathematical Sciences Department of Computer Science and Information Technology La Trobe University Melbourne Australia School of Electrical and Data Engineering University of Technology Sydney SydneyNSW2007 Australia

By executing offloaded tasks from mobile users, edge computing augments mobile devices with computing/communications resources from edge nodes (ENs), thus enabling new services/applications (e.g., real-time gaming, virtual/augmented reality). However, despite being more resourceful than mobile devices, allocating ENs’ computing/communications resources to a given favorable set of users (e.g., closer to edge nodes) may block other devices from their services. This is often the case for most existing task offloading and resource allocation approaches that only aim to maximize the network social welfare or minimize the total energy consumption but do not consider the computing/battery status of each mobile device. This work develops an energy-based proportionally fair task offloading and resource allocation framework for a multi-layer cooperative edge computing network to serve all user equipments (UEs) while considering both their service requirements and individual energy/battery levels. The resulting optimization involves both binary (offloading decisions) and continuous (resource allocation) variables. To tackle the NP-hard mixed integer optimization problem, we leverage the fact that the relaxed problem is convex and propose a distributed algorithm, namely the dynamic branch-and-bound Benders decomposition (DBBD). DBBD decomposes the original problem into a master problem (MP) for the offloading decisions and multiple subproblems (SPs) for resource allocation. To quickly eliminate inefficient offloading solutions, the MP is integrated with powerful Benders cuts exploiting the ENs’ resource constraints. We then develop a dynamic branch-and-bound algorithm (DBB) to efficiently solve the MP considering the load balance among ENs. The SPs can either be solved for their closed-form solutions or be solved in parallel at ENs, thus reducing the complexity. The numerical results show that the DBBD returns the optimal solution in maximizing the proportional fairness among U

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

BookChain: A Secure Library Book Storing and Sharing in Academic Institutions using Blockchain Technology

BookChain: A Secure Library Book Storing and Sharing in Acad...

引用

International Conference on computer Communication and Informatics (ICCCI)

作者： P. Chinnasamy Anupriya Elumalai Ramesh Kumar Ayyasamy S.P. Kavya S. Dhanasekaran Ajmeera Kiran Department of Computer Science and Engineering MLR Institute of Technology Hyderabad India School of Computer Science and Engineering VIT – AP University Faculty of Information and Communication Technology Universiti Tunku Abdul Rahman (UTAR) Kampar Malaysia Department of Artificial Intelligence and Data Science KPR Institute of Engineering and Technology Coimbatore India Department of Information Technology Kalasalingam Academy of Research and Education Deemed to be University Srivilliputtur Tamilnadu India

By exploiting wireless operators, contemporary bookcrossing allows people to share books conveniently and hastens the spread of knowledge. Nevertheless, the widespread use of smartphone bookcrossing has been seriously hampered by the lack of tracking, resulting in missing and disappearance of textbooks, additionally, it leads to organizational inefficiency. Hence, we suggest using the emerging blockchain technology to exploit the advantages of its immutability and encrypted smart contract in order to overcome these difficulties. In this research, we propose a BookChain, a blockchain-based intra-campus book-sharing system that is verifiable and effective. BookChain securely records all of the book’s interacting entities on blockchain because every consumer could retrace the book’s renting activity, hence lowering the risk of a book being misplaced. Additionally, BookChain promotes the use of smart contracts to streamline the distribution of books with very little human interference, maximizing productivity. The experimental findings demonstrate the efficiency and affordability of the suggested system even with a large number of concurrent participants.

关键词： Wireless communication Productivity Smart contracts Force Interference Decentralized applications Libraries

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：