检索结果-内蒙古大学图书馆

38th Conference on Neural information Processing Systems, NeurIPS 2024

作者： Jena, Rohit Sethi, Deeksha Chaudhari, Pratik Gee, James C. Computer and Information Science United States Electrical and Systems Engineering United States Radiology United States Penn Image Computing and Science Laboratory United States

Classical optimization and learning-based methods are the two reigning paradigms in deformable image registration. While optimization-based methods boast generalizability across modalities and robust performance, learning-based methods promise peak performance, incorporating weak supervision and amortized optimization. However, the exact conditions for either paradigm to perform well over the other are shrouded and not explicitly outlined in the existing literature. In this paper, we make an explicit correspondence between the mutual information of the distribution of per-pixel intensity and labels, and the performance of classical registration methods. This strong correlation hints to the fact that architectural designs in learning-based methods is unlikely to affect this correlation, and therefore, the performance of learning-based methods. This hypothesis is thoroughly validated with state-of-the-art classical and learning-based methods. However, learning-based methods with weak supervision can perform high-fidelity intensity and label registration, which is not possible with classical methods. Next, we show that this high-fidelity feature learning does not translate to invariance to domain shift, and learning-based methods are sensitive to such changes in the data distribution. We reassess and recalibrate performance expectations from classical and DLIR methods under access to label supervision, training time, and its generalization capabilities under minor domain shifts. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep learning in medical image registration: magic or mirage? 24

Deep learning in medical image registration: magic or mirage...

引用

Proceedings of the 38th International Conference on Neural information Processing Systems

作者： Rohit Jena Deeksha Sethi Pratik Chaudhari James C. Gee Computer and Information Science and Penn Image Computing and Science Laboratory Computer and Information Science Computer and Information Science and Electrical and Systems Engineering Computer and Information Science and Radiology and Penn Image Computing and Science Laboratory

ISBN: (纸本)9798331314385

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep Learning in Medical image Registration: Magic or Mirage?

arXiv

引用

arXiv 2024年

作者： Jena, Rohit Sethi, Deeksha Chaudhari, Pratik Gee, James C. Computer and Information Science Electrical and Systems Engineering Radiology Penn Image Computing and Science Laboratory United States

Classical optimization and learning-based methods are the two reigning paradigms in deformable image registration. While optimization-based methods boast generalizability across modalities and robust performance, learning-based methods promise peak performance, incorporating weak supervision and amortized optimization. However, the exact conditions for either paradigm to perform well over the other are shrouded and not explicitly outlined in the existing literature. In this paper, we make an explicit correspondence between the mutual information of the distribution of per-pixel intensity and labels, and the performance of classical registration methods. This strong correlation hints to the fact that architectural designs in learning-based methods is unlikely to affect this correlation, and therefore, the performance of learning-based methods. This hypothesis is thoroughly validated with state-of-the-art classical and learning-based methods. However, learning-based methods with weak supervision can perform high-fidelity intensity and label registration, which is not possible with classical methods. Next, we show that this high-fidelity feature learning does not translate to invariance to domain shift, and learning-based methods are sensitive to such changes in the data distribution. Finally, we propose a general recipe to choose the best paradigm for a given registration problem, based on these observations. © 2024, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Online Queue-Aware Service Migration and Resource Allocation in Mobile Edge computing

引用

IEEE Transactions on Vehicular Technology 2025年第5期74卷 8063-8078页

作者： Du, An Jia, Jie Chen, Jian Wang, Xingwei Huang, Min Northeastern University School of Computer Science and Engineering The Engineering Research Center of Security Technology of Complex Network System Shenyang110819 China Northeastern University Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Shenyang110819 China Northeastern University College of Information Science and Engineering State Key Laboratory of Synthetical Automation for Process Industries Shenyang110819 China

Mobile edge computing (MEC) integrated with Network Functions Virtualization (NFV) helps run a wide range of services implemented by Virtual Network Functions (VNFs) deployed at MEC networks. This emerging paradigm offers flexible edge resource management for Internet Service Providers (ISPs) and improves service satisfaction of diverse applications. However, a critical challenge lies in processing extremely unpredictable and bursty traffic flow, especially in event-trigger sensing tasks. In this paper, we study queue-aware service migration and resource allocation in an MEC network, where different devices with different arrival patterns request different network services. To this end, we first formulate a novel long-term operation cost minimization problem subject to the stability of multiple queues by adopting a simultaneous migration scheme and determining resource allocation dynamically. To address the problem efficiently, we propose an online control framework to make decisions without any prior system *** first adopt the Lyapunov optimization technique to decompose the long-term optimization problem into multiper-slot subproblems. Then, we adopt a decoupled optimization algorithm based on the coordinate descent method and augmented Lagrangian method to solve the offline mixed integer non-linear programming (MINLP) subproblems. Extensive results reveal that the proposed online optimization framework can efficiently balance queue stability and cost consumption. © 1967-2012 IEEE.

关键词： Mobile edge computing

来源：评论

学校读者我要写书评

暂无评论

Adaptive delay-energy balanced partial offloading strategy in Mobile Edge computing networks

引用

Digital Communications and Networks 2023年第6期9卷 1310-1318页

作者： Shumei Liu Yao Yu Lei Guo Phee Lep Yeoh Branka Vucetic Yonghui Li School of Computer Science and Engineering Northeastern UniversityShenyang110819China Key Laboratory of Intelligent Computing in Medical Image Ministry of EducationNortheastern UniversityShenyang110819China School of Communication and Information Engineering Chongqing University of Posts and TelecommunicationsChongqing400065China School of Electrical and Information Engineering University of SydneySydneyNSW2006Australia

Mobile Edge computing(MEC)-based computation offloading is a promising application paradigm for serving large numbers of users with various delay and energy *** this paper,we propose a flexible MECbased requirement-adaptive partial offloading model to accommodate each user's specific preference regarding delay and energy *** address the dimensional differences between time and energy,we introduce two normalized parameters and then derive the computational overhead of processing *** from existing works,this paper considers practical variations in the user request patterns,and exploits a flexible partial offloading mode to minimize computation overheads subject to tolerable delay,task workload and power *** the resulting problem is non-convex,we decouple it into two convex subproblems and present an iterative algorithm to obtain a feasible offloading *** experiments show that our proposed scheme achieves a significant improvement in computation overheads compared with existing schemes.

关键词： Mobile edge computing(MEC) Delay Energy consumption Dynamic balance Partial computation offloading

来源：评论

学校读者我要写书评

暂无评论

OH-DRL: An AoI-Guaranteed Energy-Efficient Approach for UAV-Assisted IoT Data Collection

引用

IEEE Transactions on Wireless Communications 2025年第6期24卷 5009-5022页

作者： Yang, Bowen Yu, Yao Hao, Xin Yeoh, Phee Lep Zhang, Junxiong Guo, Lei Li, Yonghui Northeastern University School of Computer Science and Engineering Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Shenyang110819 China Deakin University School of Information Technology VIC Melbourne3216 Australia University of the Sunshine Coast School of Science Technology and Engineering QLD Sunshine Coast4556 Australia Chongqing University of Posts and Telecommunications School of Communication and Information Engineering Chongqing400065 China The University of Sydney School of Electrical and Computer Engineering SydneyNSW2006 Australia

In this paper, we propose a hierarchical optimization approach that guarantees the maximum age of information (AoI) for uncrewed aerial vehicle (UAV) assisted Internet-of-Things (IoT) data collection. Our model is based on an energy-efficient simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) beamforming model. We formulate the optimization to minimize the UAV flight energy consumption subject to a maximum average AoI threshold by optimizing the UAV trajectory, IoT device scheduling, and STAR-RIS beamforming. To solve this, we develop an optimization-based hierarchical deep reinforcement learning (OH-DRL) algorithm that decomposes the formulated problem into an inter-cluster UAV visiting policy and STAR-RIS-based intra-cluster IoT scheduling policy. In OH-DRL, we jointly optimize the two policies in a high-level loop and a low-level loop, respectively. In the high-level loop, we design an AoI-guided DRL algorithm to determine the AoI-guaranteed UAV hovering position with minimal flight distance. In the low-level loop, a semidefinite relaxation (SDR)-based optimization algorithm further reduces the UAV’s flying time by minimizing the average AoI. Simulation results validate that OH-DRL achieves better convergence performance and energy-saving efficiency across different network scales. Compared to the state-of-the-art DRL algorithm, OH-DRL reduces the UAV flight energy consumption by 14.4% and decreases the number of training episodes required for convergence by 66% © 2002-2012 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Automatic skin lesion classification using a novel densely connected convolutional network integrated with an attention module

引用

Journal of Ambient Intelligence and Humanized computing 2023年第7期14卷 8943-8956页

作者： Shan, Pufang Chen, Jialei Fu, Chong Cao, Lin Tie, Ming Sham, Chiu-Wing School of Computer Science and Engineering Northeastern University Shenyang110819 China Engineering Research Center of Security Technology of Complex Network System Ministry of Education Shenyang China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Northeastern University Shenyang110819 China School of Information and Communication Engineering Beijing Information Science and Technology University Beijing100101 China Science and Technology on Space Physics Laboratory Beijing100076 China School of Computer Science The University of Auckland Auckland New Zealand

Automated skin lesion classification in dermoscopy images remains challenging due to the existence of artefacts and intrinsic cutaneous features, diversity of lesion morphology, insufficiency of training data, and class imbalance problem. To address these challenges, we propose a new densely connected convolutional network termed AttDenseNet-121, which is obtained by integrating the convolutional block attention module (CBAM) into DenseNet-121. CBAM is a simple yet efficient attention module, which is further improved by adding a novel pooling layer (multiscale-pooling) to effectively enhance its attention capability. The optimized CBAM strengthens the representation power of DenseNet-121 by emphasizing meaningful features and suppressing unnecessary ones, thus significantly enhancing the classification performance. Besides, to handle the imbalanced dataset, we employ an improved focal loss function rather than the traditional cross-entropy loss function to train AttDenseNet-121. The improved focal loss is calculated from a different perspective compared with the original focal loss. It makes the distribution of positive samples and negative samples in each batch the same as that in the original data, successfully mitigating the negative influence of class imbalance on this multi-class classification task. We conduct extensive experiments on the public benchmark dataset (HAM10000) and the results indicate that our proposed method achieves superior performance compared with that of the baselines and state-of-the-art algorithms. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Competition on robust deep learning

引用

National science Review 2023年第6期10卷 13-15页

作者： Yinpeng Dong Chang Liu Wenzhao Xiang Hang Su Jun Zhu Department of Computer Science and Technology Institute for AI Tsinghua-Bosch Joint ML Center THBI Lab BNRist Center Tsinghua University Institute of Image Communication and Networks Engineering in the Department of Electronic Engineering (EE) Shanghai Jiao Tong University Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Peng Cheng Laboratory Pazhou Laboratory (Huangpu)

PROBLEM In recent years,the rapid development of artificial intelligence (AI) technology,especially machine learning and deep learning, is profoundly changing human production and *** various fields,such as robotics,face recognition,autonomous driving and healthcare,AI is playing an important ***,although AI is promoting the technological revolution and industrial progress,its security risks are often *** studies have found that the wellperforming deep learning models are extremely vulnerable to adversarial examples [1-3].The adversarial examples are crafted by applying small,humanimperceptible perturbations to natural examples,but can mislead deep learning models to make wrong *** vulnerability of deep learning models to adversarial examples can raise security and safety threats to various realworld applications.

关键词： robustness adversarial example adversarial training deep learning

来源：评论

学校读者我要写书评

暂无评论

An AoI-Guaranteed Sensor Data Collection Strategy for RIS-assisted UAV Communication System

An AoI-Guaranteed Sensor Data Collection Strategy for RIS-as...

引用

2023 IEEE/CIC International Conference on Communications in China, ICCC 2023

作者： Yang, Bowen Yu, Yao Li, Jianqi Xu, Tongze Deng, Der-Jiunn Chin, Hui-Hsin Li, Yonghui Northeastern University School of Computer Science and Engineering Shenyang China National Changhua University of Education Department of Computer Science and Information Engineering China Overseas Chinese University Department of Information Technology China University of Sydney School of Electrical and Information Engineering Sydney Australia Northeastern University Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Shenyang China

ISBN: (纸本)9798350345384

The recently proposed concept of Age of information (AoI) measures the freshness of the sensor data sampled by remote Internet of Things (IoT) devices, which is an important indicator for the smart city. Unmanned Aerial Vehicles (UAV) communication can significantly reduce the sensor data's AoI due to its high maneuverability. However, the UAV's high energy consumption remains a challenging issue. In this paper, we propose a Reconfigurable Intelligent Surface (RIS) assisted UAV data collection framework to minimize UAV energy consumption. By adjusting the phase of the RIS, the UAV can collect more sensor data simultaneously. A RIS phase alignment method is introduced to reduce the complexity of RIS phase solving. The IoT devices are clustered to discretize UAV trajectory. The optimal clustering scheme to maximize the AoI suppression efficiency is found by adjusting the RIS phase alignment points continuously. Therefore, the AoI requirement can be achieved with a shorter UAV flight distance. Finally, the UAV trajectory is solved by the Deep Q network (DQN) algorithm to guarantee the long-term AoI requirement. Simulation results validate the effectiveness of the proposed method. © 2023 IEEE.

关键词： Maneuverability

来源：评论

学校读者我要写书评

暂无评论

Cross-Domain Retinopathy Classification with OCT images via Disentangling Representation and Adaptation Networks

Cross-Domain Retinopathy Classification with OCT Images via ...

引用

Optoelectronics Global Conference (OGC)

作者： Wickramaarachchi Minidu Thiranjaya Yuan Li Lei Tao Yuemei Luo Jiangsu Provincial Key Laboratory of Intelligent Medical Image Computing (iMIC) School of Artificial Intelligence Nanjing University of Information Science and Technology Nanjing China Jiangsu Provincial Key Laboratory of Intelligent Medical Image Computing (iMIC) School of Computer Science Nanjing University of Information Science and Technology Nanjing China Jiangsu Provincial Key Laboratory of Intelligent Medical Image Computing (iMIC) School of Future Technology Nanjing University of Information Science and Technology Nanjing China

ISBN: (数字)9798350391466

ISBN: (纸本)9798350391473

Deep learning methods have shown significant potential in retinopathy classification using optical coherence tomography (OCT) images. However, substantial challenges arise due to domain shift issues, which are attributable to heterogeneity in data collection devices, variations in subject populations, and discrepancies in imaging protocols and parameters. To address these, we propose a novel approach leveraging Disentangling Representation and Adaptation Networks (DRANet) to enhance the robustness and generalization of retinopathy classification models across different OCT datasets. Our methodology involves developing DRANet, a specialized network architecture designed to disentangle representations of data and facilitate domain adaptation. DRANet includes components for learning invariant representations across domains through adversarial training, ensuring robustness against domain shifts. DRANet encodes individual representations of content and style from both source and target images. Then, it adapts the domain by incorporating the transferred style factor into the content factor, along with learnable weights specified for each domain. This learning framework allows bidirectional domain. After conducting extensive experiments conducted on three public datasets achieved accuracies of $\mathrm{8 8 \%, ~} \mathrm{9 0 \%}$, and $\mathrm{9 2 \%}$ under three cross-domain scenarios, demonstrating the efficacy of our approach in mitigating domain shift effects and improving retinopathy classification performance.

关键词： Training Deep learning Protocols Retinopathy Optical coherence tomography Imaging Coherence Network architecture Data collection Robustness

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：