检索结果-内蒙古大学图书馆

IEEE International Symposium on Biomedical Imaging

作者： Jiarui Xing Nian Wu Kenneth C. Bilchick Frederick H. Epstein Miaomiao Zhang Department of Electrical and Computer Engineering University of Virginia USA School of Medicine University of Virginia Health System USA Department of Biomedical Engineering University of Virginia Health System USA Department of Computer Science University of Virginia USA

ISBN: (数字)9798350313338

ISBN: (纸本)9798350313345

Class imbalance in the medical image dataset is almost inherent due to the limited availability of clinical data for certain diseases and patient populations. Under-represented classes in the training set affect the classification task because the classifier tends to learn more from the majority classes, which are more common in the dataset and ignore data from the minority classes. To mitigate this issue, we propose a method to learn using two different convolutional neural network-based experts; such experts try to learn boundaries within the head classes, between the head and tail classes, and within the tail classes. During expert training, we integrate the MixUp regularization method to augment imbalanced data, employing distinct data sampling strategies for more effective mixing compared to random selection in traditional MixUp. During the inference phase, we combine the logits of the different experts based on their expertise in the corresponding classes. This way, we can improve the accuracy of the head and tail classes. Experiments using highly imbalanced and long-tailed datasets demonstrate the effectiveness of the suggested framework.

关键词： Training Accuracy Head Three-dimensional displays Tail Data models Task analysis

来源：评论

学校读者我要写书评

暂无评论

An Improved YOLO-v3Algorithm for Ship Detection in SAR Image Based on K-means++ with Focal Loss 3

An Improved YOLO-v3Algorithm for Ship Detection in SAR Image...

引用

3rd China International SAR Symposium, CISS 2022

作者： Wang, Haonan Wu, Baolong Wu, Yanni Zhang, Shuangxi Mei, Shaohui Liu, Yanyang Northwestern Polytechnical University School of Electronics and Information Xi'an China Shanxi Normal University School of Computer Science Xi'an China Shanghai Institute of Satellite Engineering Department of Radar System Engineering Shanghai China

ISBN: (数字)9798350398717

ISBN: (纸本)9798350398717

In recent years, practical industrial production application have put forward extremely high requirements for its detection accuracy and detection efficiency in the aspect of synthetic aperture radar (SAR) image ship detection. Among the solutions to this problem, the YOLO has received more and more attention due to its advantages such as high speed. In this paper, the K-means++ is used to obtain the Anchor Box, the Focal loss is introduced to balance the proportion of positive and negative samples, and an improved image detection algorithm based on YOLO-v3 is proposed to solve the low detection efficiency and detection accuracy of ship images. The experimental results show that the improved algorithm in this paper can get rid of the local optimum well, shorten the convergence time, improve the training efficiency and the accuracy of ship image detection. © 2022 IEEE.

关键词： Ships

来源：评论

学校读者我要写书评

暂无评论

PipeEdge: A Trusted Pipelining Collaborative Edge Training based on Blockchain 23

PipeEdge: A Trusted Pipelining Collaborative Edge Training b...

引用

2023 World Wide Web Conference, WWW 2023

作者： Yuan, Liang He, Qiang Chen, Feifei Dou, Ruihan Jin, Hai Yang, Yun School of Computer Science and Technology Huazhong University of Science and Technology China Department of Computing Technologies Swinburne University of Technology Australia School of Information Technology Deakin University Australia Faculty of Mathematics University of Waterloo Canada National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9781450394161

Powered by the massive data generated by the blossom of mobile and Web-of-Things (WoT) devices, Deep Neural Networks (DNNs) have developed both in accuracy and size in recent years. Conventional cloud-based DNN training incurs rapidly-increasing data and model transmission overheads as well as privacy issues. Mobile edge computing (MEC) provides a promising solution by facilitating DNN model training on edge servers at the network edge. However, edge servers often suffer from constrained resources and need to collaborate on DNN training. Unfortunately, managed by different telecoms, edge servers cannot properly collaborate with each other without incentives and trust. In this paper, we introduce PipeEdge, a scheme that promotes collaborative edge training between edge servers by introducing incentives and trust based on blockchain. Under the PipeEdge scheme, edge servers can hire trustworthy workers for pipelined DNN training tasks based on model parallelism. We implement PipeEdge and evaluate it comprehensively with four different DNN models. The results show that it outperforms state-of-the-art schemes by up to 173.98% with negligible overheads. © 2023 ACM.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

HairDiffusion: vivid multi-colored hair editing via latent diffusion 24

HairDiffusion: vivid multi-colored hair editing via latent d...

引用

Proceedings of the 38th International Conference on Neural Information Processing systems

作者： Yu Zeng Yang Zhang Jiachen Liu Linlin Shen Kaijun Deng Weizhao He Jinbao Wang Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University and Shenzhen Institute of Artificial Intelligence and Robotics for Society and National Engineering Laboratory for Big Data System Computing Technology Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen University and Guangdong Provincial Key Laboratory of Intelligent Information Processing

ISBN: (纸本)9798331314385

Hair editing is a critical image synthesis task that aims to edit hair color and hairstyle using text descriptions or reference images, while preserving irrelevant attributes (e.g., identity, background, cloth). Many existing methods are based on StyleGAN to address this task. However, due to the limited spatial distribution of StyleGAN, it struggles with multiple hair color editing and facial preservation. Considering the advancements in diffusion models, we utilize Latent Diffusion Models (LDMs) for hairstyle editing. Our approach introduces Multi-stage Hairstyle Blend (MHB), effectively separating control of hair color and hairstyle in diffusion latent space. Additionally, we train a warping module to align the hair color with the target region. To further enhance multi-color hairstyle editing, we fine-tuned a CLIP model using a multi-color hairstyle dataset. Our method not only tackles the complexity of multi-color hairstyles but also addresses the challenge of preserving original colors during diffusion editing. Extensive experiments showcase the superiority of our method in editing multi-color hairstyles while preserving facial attributes given textual descriptions and reference images.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Contrastive Learning-Based Generic Audio-to-Lyrics Alignment 21

Contrastive Learning-Based Generic Audio-to-Lyrics Alignment

引用

21st IEEE International Symposium on Parallel and Distributed Processing with Applications, 13th IEEE International Conference on Big Data and Cloud Computing, 16th IEEE International Conference on Social Computing and Networking and 13th International Conference on Sustainable Computing and Communications, ISPA/BDCloud/SocialCom/SustainCom 2023

作者： Jin, Zhenkun Huang, Shuai Nie, Xin Zhou, Xinlei Yi, Yuanyuan Zhou, Gefei Wuhan Business UIniversity Wuhan China Wuhan Research Institute of Posts and Telecommunications Wuhan China Hubei Key Laboratory of Distributed System Security Wuhan China Hubei Engineering Research Center on BigData Security Wuhan China Huazhong University of Science and Technology School of Cyber Science and Engineering Wuhan China Huazhong University of Science and Technology School of Journalism and Information Communication Wuhan China Huazhong University of Science and Technology School of Computer Science and Technology Wuhan China

ISBN: (纸本)9798350329223

In this research, we tackled the complex task of aligning audio and lyrics content automatically. This task entails the precise matching of lyrics with the corresponding audio segments in songs, necessitating the coordination of two distinct modalities: text and audio. To surmount this challenge, we employed a contrastive learning approach that generates cross-modal embeddings capable of establishing connections between the audio and text domains. Furthermore, we adopted a hierarchical and cascading design. Initially, the system predicts synchronization at the sentence level and subsequently refines the alignment at the word level. This design equips the system to handle extended sequences effectively, as cross-correlations leverage secondary memory that corresponds to sequence length. Through this innovative approach, we developed a trainable end-to-end system that can harness weakly annotated training data, concurrently learning a robust text model optimized for alignment tasks. Our experiments demonstrated substantial enhancements in average alignment accuracy when utilizing our proposed system, underscoring its robustness compared to previous state-of-the-art models. © 2023 IEEE.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Formation and Obstacle Avoidance Control Based on Multi-Agent Reinforcement Learning

引用

IEEE Transactions on Network science and engineering 2025年

作者： Niu, Fuxi Nian, Xiaohong Pan, Chao Dia, Xunhua Wang, Haibo Xiong, Hongyun Central South University Key Lab of Institute of cluster unmanned system School of automation Changsha410083 China Central South University School of computer science and engineering Changsha410083 China

For the multi-agent formation control problem, the current mainstream control methods are based on graph theory and consistency theory, which often require precise modeling of the system and rigorous mathematical reasoning. Based on the MADDPG deep reinforcement learning algorithm, this paper models the formation control problem as a reinforcement learning problem by considering various constraints on the agent, and obtains the behavior strategy of each agent through learning and training. Furthermore, this paper combines the leader-follow method to divide all agents into two categories, and solves the problem that it is difficult for agents to balance the long-term rewards of formation and obstacle avoidance tasks in reinforcement learning methods. The simulation experiments are divided into two types: circular formation control and formation obstacle avoidance. The dynamic movement process of the agent in the corresponding experimental scene is shown, and the effectiveness of the algorithm is verified by combining the formation process error. © 2025 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Simultaneous Wireless Information and Power Transfer-based Multi-hop Clustering Routing Protocol for RF Energy Harvesting-Cognitive Radio Sensor Networks

Journal of Network Intelligence

引用

Journal of Network Intelligence 2023年第3期8卷 991-1007页

作者： Wang, Ji-Hong Su, Xi-Tong Ge, Yi-Yang Chen, Yeh-Cheng Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology Ministry of Education Northeast Electric Power University China School of Electrical Engineering Northeast Electric Power University 169 Changchun Road Jilin132012 China Department of Computer Science University of California DavisCA001313 United States

Radio frequency energy harvesting (RF EH) and clustering are promising solutions to solve the energy constraint problem faced by cognitive radio sensor networks (CRSNs). However, study on clustering routing protocol design for RF EH-CRSNs is still in the initial stage, and there lacks of solutions for inter-cluster routing problem, which restricts the network scalability. In addition, unbalanced harvested energy and energy consumption may result in premature death of nodes in key position, which will significantly reduce the network performance. In this paper, a simultaneous wireless information and power transfer (SWIPT)-based multi-hop clustering routing protocol (SMCRP) is proposed for RF EH-CRSNs to well balance the residual energy distribution among nodes and postpone node death. Apart from the RF energy harvested from dedicated or ambi-ent energy sources, the energy obtained through intra-cluster and inter-cluster SWIPT is also considered to facilitate high-quality cluster heads selection, cluster formation and inter-cluster route establishment. Power splitting ratio is derived based on the residual energy of CRSNs nodes and their energy consumption to enable effective implementation of SWIPT in CRSNs. Simulation results show that SMCRP is effective in balancing the residual energy distribution among nodes and improving node sustainability. To be specific, compared with current clustering protocols, the network lifetime is extended by 64.07% while intra-cluster energy balance degree and network energy balance degree are improved by 15.1% and 7.82%, respectively. © 2023 ISSN 2414-8105 (Online).

关键词： Cognitive radio

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Overview of Multi-agent systems for Controlling Smart Grids

引用

CSEE Journal of Power and Energy systems 2022年第1期8卷 115-131页

作者： Om Prakash Mahela Mahdi Khosravy Neeraj Gupta Baseem Khan Hassan Haes Alhelou Rajendra Mahla Nilesh Patel Pierluigi Siano Power System Planning Division Rajasthan Rajya Vidyut Prasaran Nigam Ltd.Jaipur-302005India Media Integrated Communication Laboratory Graduate School of EngineeringOsaka UniversityOsakaJapan Computer Science and Engineering Department Oakland UniversityRochesterUSA Department of Electrical and Computer Engineering Hawassa UniversityHawassa 1530Ethiopia Faculty of Mechanical and Electrical Engineering Tishreen UniversityLattakiaSyria Department of Electrical Engineering National Institute of Technology Kurukshetra Department of Management&Innovation Systems University of Salerno84084 SalernoItaly

Agents are intelligent entities that act flexibly and autonomously and make wise decisions based on their intelligence and experience.A multi-agent system(MAS)contains multiple,intelligent,and interconnected collaborating agents for solving a problem beyond the ability of a single agent.A smart grid(SG)combines advanced intelligent systems,control techniques,and sensing methods with an existing utility power *** controlling smart grids,various control systems with different architectures have already been ***-based control of power system operations has been shown to overcome the limitations of time required for analysis,relaying,and protection;transmission switching;communication protocols;and management of plant *** systems provide an alternative for fast and accurate power network *** paper provides a comprehensive overview of MASs used for the control of smart *** paper provides a wide-spectrum view of the status of smart grids,MAS-based control techniques and their implementation for the control of smart *** of MASs in the control of various aspects of smart grids-including the management of energy,marketing energy,pricing,scheduling energy,reliability,network security,fault handling capability,communication between agents,SG-electrical vehicles,SG-building energy systems,and soft grids—have been critically *** than a hundred publications on the topic of MAS-based control of smart grids have been critically examined,classified,and arranged for fast reference.

关键词： Coordinated control multi-agent systems renewable energy sources smart energy infrastructure smart grid

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Techniques Using the Rapid Miner Tool for Solar Production Forecasting 4

Machine Learning Techniques Using the Rapid Miner Tool for S...

引用

4th International Conference on Digital Age and Technological Advances for Sustainable Development, ICDATA 2023

作者： Ledmaoui, Younes El Fahli, Asmaa Elmaghraoui, Adila El Aroussi, Mohamed Saadane, Rachid Chehri, Abdellah Hassania School of Public Works Casablanca Laboratory Engineering System Morocco Mohammed Vi Polytechnic University Benguerir Morocco Mathematics and Computer Science KingstonONK7K 7B4 Canada

ISBN: (纸本)9798350311112

The Artificial Neural Network (ANN), Extreme Gradient Boosting (XGBOOST), Support Vector Machine (SVM), and Random Forest (RF) machine learning (ML) algorithms are used in this research to give a comparative study of solar energy production forecasting in Morocco. The models were developed, trained, and then assessed. In this paper, two metrics were employed to evaluate the models' performance: Mean absolute error (MAE), and root mean squared error (RMSE). The models' performances show that ANN is the most reliable predictive model for energy forecasting in similar scenarios. © 2023 IEEE.

关键词： Solar energy

来源：评论

学校读者我要写书评

暂无评论

Deepfake Detection and Mitigation Using Advanced CNN: Ensuring Digital Content Integrity

Deepfake Detection and Mitigation Using Advanced CNN: Ensuri...

引用

Advancement in Computation & computer Technologies (InCACCT), International Conference on

作者： Palvi Sharma Shubham Gupta Monu Sharma Neha Gupta Dept. of CSE Model Institute of Engineering & Technology Jammu India Valley Health System Winchester Virginia School of Computer Science and Engineering Geeta University

ISBN: (数字)9798331543891

ISBN: (纸本)9798331543907

Deepfake technology, with its hyper-realistic fake content and backed by AI technologies, presents a serious threat to the authenticity and security of digital content. The study deployed three different machine learning-based models: Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Vision Transformers (ViT) for discrimination of real and fake faces. The model leverages the capabilities of CNN for automatic spatial feature learning from raw data, giving it exceptional performance in the discrimination between real and fake content. Performance measures were made in terms of accuracy, precision, recall, and F1 score and compared against traditional methods; the model remains a trustworthy contribution towards deepfake detection. Findings indicate that the CNN exceeded others with the highest precisions of up to 92.21%, recall of 90%, F1 score of 93%, and perfect accuracy of 97.25%. The paper further discusses the future developments, for instance, architectural improvements and real-time applications and expanding detection capacity outside visual data for a more formidable approach towards new deepfake formats, such as audio and text. This endeavor lays the groundwork for viable and scalable roads against the rising onslaught on digital media manipulation, ensuring content integrity in an AI-driven world.

关键词： Support vector machines Representation learning Deepfakes Visualization Accuracy Roads Transformers Convolutional neural networks Security Artificial intelligence

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：