检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Lin, Yuqi Chen, Minghao Zhang, Kaipeng Li, Hengjia Li, Mingming Yang, Zheng Lv, Dongqin Lin, Binbin Liu, Haifeng Cai, Deng State Key Lab of CAD&CG College of Computer Science Zhejiang University China Hangzhou Dianzi University China Shanghai Artificial Intelligence Laboratory China FABU Inc. School of Software Technology Zhejiang University China Nantong Port Group China

Contrastive Language-Image Pre-training (CLIP) has demonstrated impressive capabilities in open-vocabulary classification. The class token in the image encoder is trained to capture the global features to distinguish different text descriptions supervised by contrastive loss, making it highly effective for single-label classification. However, it shows poor performance on multi-label datasets because the global feature tends to be dominated by the most prominent class and the contrastive nature of softmax operation aggravates it. In this study, we observe that the multi-label classification results heavily rely on discriminative local features but are overlooked by CLIP. As a result, we dissect the preservation of patch-wise spatial information in CLIP and proposed a local-to-global framework to obtain image tags. It comprises three steps: (1) patch-level classification to obtain coarse scores;(2) dual-masking attention refinement (DMAR) module to refine the coarse scores;(3) class-wise reidentification (CWR) module to remedy predictions from a global perspective. This framework is solely based on frozen CLIP and significantly enhances its multi-label classification performance on various benchmarks without dataset-specific training. Besides, to comprehensively assess the quality and practicality of generated tags, we extend their application to the downstream task, i.e., weakly supervised semantic segmentation (WSSS) with generated tags as image-level pseudo labels. Experiments demonstrate that this classify-then-segment paradigm dramatically outperforms other annotation-free segmentation methods and validates the effectiveness of generated tags. Our code is available at https://***/linyq2117/TagCLIP. Copyright © 2023, The Authors. All rights reserved.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Defense Against Adversarial Faces at the Source: Strengthened Faces Based on Hidden Disturbances

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年

作者： Li, Shuangliang Wang, Jinwei Wu, Hao Zhang, Jiawei Cheng, Xin Luo, Xiangyang Ma, Bin School of Software Nanjing Universityof Information Science and Technology Nanjing210044 China College of Cryptology and Cyber Science Nankai University Tianjing300073 China State Key Laboratory of Mathemat-ical Engineering and Advanced Computing Zhengzhou450002 China School of Cyber Security and Information Law Chongqing University of Posts and Telecommunications Chongqing400065 China School of Computer Science Nanjing University of Information Science and Technology Nanjing210044 China School of Cyber Security Qilu University of Technology Shandong Academy of Sciences Shandong250353 China

Face Recognition (FR) systems, while widely used across various sectors, are vulnerable to adversarial attacks, particularly those based on deep neural networks. Despite existing efforts to enhance the robustness of FR models, they still face the risk of secondary adversarial attacks. To address this, we propose a novel approach employing "strengthened face" with preemptive defensive perturbations. Strengthened face ensures original recognition accuracy while safeguarding FR systems against secondary attacks. In the white-box scenario, the strengthened face utilizes gradient-based and optimization-based methods to minimize feature representation differences between face pairs. For the black-box scenario, we propose Shielded Gradient Sign Descent (SGSD) to optimize the gradient update direction of strengthened faces, ensuring the transferability and effectiveness against unknown adversarial attacks. Experimental results demonstrate the efficacy of strengthened faces in defending against adversarial faces without compromising the performance of FR models or face image visual quality. Moreover, SGSD outperforms conventional methods, achieving an average performance improvement of 4% in transferability across different attack intensities. © 2020 IEEE.

关键词： adversarial attacks face recognition facial protection

来源：评论

学校读者我要写书评

暂无评论

Compound-Action DRL for Timely Data Collection in UAV-Assisted IoT with Stochastic Arrival

Compound-Action DRL for Timely Data Collection in UAV-Assist...

引用

International Conference on Communication Technology (ICCT)

作者： Mengjie Yi Xijun Wang Juan Liu Yan Zhang Ronghui Hou School of Cyber Engineering Xidian University Xi'an China School of Electronics and Information Technology Sun Yat-sen University Guangzhou China School of Electrical Engineering and Computer Science Ningbo University Zhejiang China State Key Lab of Integrated Service Networks Information Science Institute Xidian University Xi'an Shaanxi China

The use of unmanned aerial vehicles (UAVs) in IoT networks can provide an efficient means of collecting mission-critical data. However, ensuring the freshness of collected data is a significant challenge due to the stochastic status updates of sensor nodes (SN s), In this paper, we delve into the issue of timely data collection aided by a UAV in IoT networks, where the UAV collects status updates from SNs and offloads the data to the data center at an opportune location. SNs sample information from their surroundings at stochastic intervals to generate new status updates. To minimize the average total age of information (AoI), we describe the problem as a Markov decision process and propose a compound-action deep reinforcement learning (CADRL) approach to jointly optimize the trajectory, scheduling of SNs, offloading strategy, and offloading power of the UAV, which can handle both continuous and discrete actions of the UAV. Simulation results indicate that our algorithm effectively reduces the AoI when contrasted with baseline methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ROBUST EARLY-LEARNING: HINDERING THE MEMORIZATION OF NOISY labELS 9

ROBUST EARLY-LEARNING: HINDERING THE MEMORIZATION OF NOISY L...

引用

9th International Conference on Learning Representations, ICLR 2021

作者： Xia, Xiaobo Liu, Tongliang Han, Bo Gong, Chen Wang, Nannan Ge, Zongyuan Chang, Yi Trustworthy Machine Learning Lab School of Computer Science The University of Sydney Australia Department of Computer Science Hong Kong Baptist University Hong Kong School of Computer Science and Engineering Nanjing University of Science and Technology China ISN State Key Laboratory School of Telecommunications Engineering Xidian University China Medical AI Group Faculty of Engineering Monash University Australia Airdoc Research Monash University Australia School of Artificial Intelligence Jilin University China

The memorization effects of deep networks show that they will first memorize training data with clean labels and then those with noisy labels. The early stopping method therefore can be exploited for learning with noisy labels. However, the side effect brought by noisy labels will influence the memorization of clean labels before early stopping. In this paper, motivated by the lottery ticket hypothesis which shows that only partial parameters are important for generalization, we find that only partial parameters are important for fitting clean labels and generalize well, which we term as critical parameters;while the other parameters tend to fit noisy labels and cannot generalize well, which we term as non-critical parameters. Based on this, we propose robust early-learning to reduce the side effect of noisy labels before early stopping and thus enhance the memorization of clean labels. Specifically, in each iteration, we divide all parameters into the critical and non-critical ones, and then perform different update rules for different types of parameters. Extensive experiments on benchmark-simulated and real-world label-noise datasets demonstrate the superiority of the proposed method over the state-of-the-art label-noise learning methods. © 2021 ICLR 2021 - 9th International Conference on Learning Representations. All rights reserved.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Learning transformer-based heterogeneously salient graph representation for multimodal remote sensing image classification

arXiv

引用

arXiv 2023年

作者： Yang, Jiaqi Du, Bo Zhang, Liangpei State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University Wuhan430079 China National Engineering Research Center for Multimedia Software Wuhan University Wuhan China Institute of Artificial Intelligence Wuhan University Wuhan China School of Computer Science Wuhan University Wuhan China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China

Data collected by different modalities can provide a wealth of complementary information, such as hyperspectral image (HSI) to offer rich spectral-spatial properties, synthetic aperture radar (SAR) to provide structural information about the Earth's surface, and light detection and ranging (LiDAR) to cover altitude information about ground elevation. Therefore, a natural idea is to combine multimodal images for refined and accurate land-cover interpretation. Although many efforts have been attempted to achieve multi-source remote sensing image classification, there are still three issues as follows: 1) indiscriminate feature representation without sufficiently considering modal heterogeneity, 2) abundant features and complex computations associated with modeling long-range dependencies, and 3) overfitting phenomenon caused by sparsely labeled samples. To overcome the above barriers, a transformer-based heterogeneously salient graph representation (THSGR) approach is proposed in this paper. First, a multimodal heterogeneous graph encoder is presented to encode distinctively non-Euclidean structural features from heterogeneous data. Then, a self-attention-free multi-convolutional modulator is designed for effective and efficient long-term dependency modeling. Finally, a mean forward is put forward in order to avoid overfitting. Based on the above structures, the proposed model is able to break through modal gaps to obtain differentiated graph representation with competitive time cost, even for a small fraction of training samples. Experiments and analyses on three benchmark datasets with various state-of-the-art (SOTA) methods show the performance of the proposed approach. Copyright © 2023, The Authors. All rights reserved.

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

Cluster-guided Contrastive Class-imbalanced Graph Classification

arXiv

引用

arXiv 2024年

作者： Ju, Wei Mao, Zhengyang Yi, Siyu Qin, Yifang Gu, Yiyang Xiao, Zhiping Shen, Jianhao Qiao, Ziyue Zhang, Ming College of Computer Science Sichuan University Chengdu China School of Computer Science State Key Laboratory for Multimedia Information Processing PKU-Anker LLM Lab Peking University Beijing China College of Mathematics Sichuan University Chengdu China Paul G. Allen School of Computer Science and Engineering University of Washington SeattleWA United States Huawei Hisilicon Shanghai China School of Computing and Information Technology Great Bay University Dongguan China

This paper studies the problem of class-imbalanced graph classification, which aims at effectively classifying the categories of graphs in scenarios with imbalanced class distribution. Despite the tremendous success of graph neural networks (GNNs), their modeling ability for imbalanced graph-structured data is inadequate, which typically leads to predictions biased towards the majority classes. Besides, existing class-imbalanced learning methods in visions may overlook the rich graph semantic substructures of the majority classes and excessively emphasize learning from the minority classes. To tackle this issue, this paper proposes a simple yet powerful approach called C3GNN that incorporates the idea of clustering into contrastive learning to enhance class-imbalanced graph classification. Technically, C3GNN clusters graphs from each majority class into multiple subclasses, ensuring they have similar sizes to the minority class, thus alleviating class imbalance. Additionally, it utilizes the Mixup technique to synthesize new samples and enrich the semantic information of each subclass, and leverages supervised contrastive learning to hierarchically learn effective graph representations. In this way, we can not only sufficiently explore the semantic substructures within the majority class but also effectively alleviate excessive focus on the minority class. Extensive experiments on real-world graph benchmark datasets verify the superior performance of our proposed method. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Challenges and countermeasures of interaction in autonomous vehicles

引用

Science China(Information Sciences) 2019年第5期62卷 5-7页

作者： Deyi LI Meng LIU Fei ZHAO Yu LIU State Key Laboratory of Software Development Environment School of Computer Science and EngineeringBeihang University Chinese Academy of Engineering School of Transportation Science and Engineering Beihang University

Autonomous vehicles can bring changes into ways of traveling, traffic, and even production modes [1].At present, the majority of researches [2] on autonomous vehicles are limited in perception, planning, decision-making and other calculation front,while interactive cognitive perspective is barely looked into. The interactive sector involved in self-driving is rather sophisticated [3]. However,

关键词： Challenges and countermeasures of interaction in autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Federated Learning for Recognizing Private Handwritten Numbers

Federated Learning for Recognizing Private Handwritten Numbe...

引用

2021 IEEE International Conference on Data Science and computer Application, ICDSCA 2021

作者： Wang, Di Jiang, Yuchen Zhao, Yibo Chen, Panyang Li, Ang Pan, Kun Institution of Information Engineering Chinese Academy of Sciences State Key Laboratory of Information Security Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Computer Science and Technology College of Computer Science Chongqing University Chongqing China Zibo Shiyan High School International Division Zibo Shandong China Computer Science Department of Computer Science and Software Engineering Auburn University Auburn United States Shanghai Shangde Experimental School Shanghai China School of Computing University of Leeds Leeds United Kingdom

ISBN: (纸本)9781665440530

Federated learning (FL) is a new technology in the field of machine learning, which trains an algorithm across multiple decentralized edge devices or servers holding data samples, without exchanging their data samples. Horizontal Federated Learning (HFL) is a FL method focusing on using the same feature space across all users, which is suitable for situations where users' data feature dimensions overlap more. Because FL has the advantage of privacy protection in the field of manufacturing, we propose an HFL-based deep learning method named FedCNN, which realizes the digits handwritten recognition without violating the privacy or data sharing. Three different methods (local model, centralized model, and HFL-based FedCNN) are used on a public data set (MINIST) to compare the pros and cons of the solutions. The experimental results demonstrate that FedCNN scheme can simultaneously achieve practical privacy preservation and good model generalization performance with Homomorphic Encryption. © 2021 IEEE.

关键词： Data privacy

来源：评论

学校读者我要写书评

暂无评论

Near Optimal Hybrid Digital-Analog Beamforming for mmWave Point-to-Point MIMO Transmissions Using OTFS Waveforms

引用

IEEE Transactions on Communications 2025年

作者： Liu, Mengmeng Li, Shuangyang Wei, Zhiqiang Bai, Baoming Caire, Giuseppe Ng, Derrick Wing Kwan Xidian University State Key Lab. of ISN Xi'an710071 China Technische Universität Berlin Electrical Engineering and Computer Science Department Berlin10587 Germany Xi'an Jiaotong University School of Mathematics and Statistics Xi'an710049 China Peng Cheng Laboratory Guangdong Shenzhen518055 China Guangdong Guangzhou510555 China University of New South Wales School of Electrical Engineering and Telecommunications SydneyNSW2052 Australia

In this paper, a point-to-point (P2P) orthogonal time frequency space (OTFS)-based multiple-input multiple-output (MIMO-OTFS) transmission scheme is devised for millimeter wave (mmWave) channels. The proposed transmission scheme relies on a low-complexity hybrid digital-analog beamforming (HBF) scheme that exploits the delay-Doppler (DD) domain channel properties, where detailed design criteria for different channel conditions are presented, including the case where paths are indistinguishable by angles. Thanks to the proposed HBF scheme, approximate path-wise interference-free transmission of multiple data streams is achieved, and consequently, only little pre-equalization is required for combating the residual channel impairments. The achievable rate of the proposed scheme is studied and compared with the orthogonal frequency-division multiplexing (OFDM) counterpart. In particular, we unveil that the condition number of the effective angular domain matrix for OTFS is smaller than that for OFDM, due to the enhanced path separability in the DD domain. As a result, the proposed MIMO-OTFS transmission scheme demonstrates superior performance over the MIMO-OFDM transmission scheme. Our numerical results corroborate our theoretical analysis and show a near-optimal rate performance with significantly reduced complexity compared to the optimal singular value decomposition (SVD) precoding method. © 1972-2012 IEEE.

关键词： Orthogonal frequency division multiplexing

来源：评论

学校读者我要写书评

暂无评论

A genetic programming method for classifier construction and cost learning in high-dimensional unbalanced classification 20

A genetic programming method for classifier construction and...

引用

2020 Genetic and Evolutionary Computation Conference, GECCO 2020

作者： Pei, Wenbin Xue, Bing Shang, Lin Zhang, Mengjie School of Engineering and Computer Science Victoria University of Wellington New Zealand State Key Laboratory for Novel Software Technology Nanjing University China

ISBN: (纸本)9781450371278

Cost-sensitive learning has been widely used to address the problem of class imbalance. However, cost matrices are often manually designed. In many real-world applications, cost values are often unknown because of the limited domain knowledge. This paper proposes a new genetic programming method to construct cost-sensitive classifiers, which do not require the manually designed cost values. The experimental results show that the proposed method often outperforms existing GP methods. © 2020 Owner/Author.

关键词： Genetic programming

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：