检索结果-内蒙古大学图书馆

Progressive Multi-Resolution Loss for Crowd Counting

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第5期34卷 3232-3244页

作者： Yan, Ziheng Qi, Yuankai Li, Guorong Liu, Xinyan Zhang, Weigang Yang, Ming-Hsuan Huang, Qingming Univ Chinese Acad Sci Sch Comp Sci & Technol Key Lab Big Data Min & Knowledge Management Beijing 101408 Peoples R China Univ Adelaide Australian Inst Machine Learning Adelaide SA 5005 Australia Harbin Inst Technol Weihai Weihai 264209 Peoples R China Univ Calif Merced EECS Merced CA 95343 USA

Crowd counting is usually handled in a density map regression fashion, which is supervised via an L2 loss between the predicted density map and ground truth. To effectively regulate models, various improved L2 loss functions have been developed to find a better correspondence between predicted density and annotation positions. In this paper, we propose to predict the density map at one resolution but measure its quality via a derived log-formed loss at multiple resolutions. Unlike existing methods that assume density maps at different resolutions are independent, our loss is obtained by modeling the likelihood function inspired by the relationship of density maps across multi-resolutions. We find that the traditional single-resolution L2 loss is a particular case of our derived log-likelihood. We mathematically prove it is superior to a single-resolution L2 loss. Without bells and whistles, the proposed loss substantially improves several baselines and performs favorably compared to state-of-the-art methods on five crowd counting datasets: NWPU-Crowd, ShanghaiTech A & B, UCF-QNRF, and JHU-Crowd++. The source code and trained models are released at https://***/streamer-AP/PML_***.

关键词： Annotations Kernel Loss measurement Bayes methods Costs Predictive models Bandwidth Crowd counting crowd analysis multi-resolution cascade loss

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study on Multi-modal Fusion for Automated Lung Disease Diagnostics 4

A Comparative Study on Multi-modal Fusion for Automated Lung...

引用

4th International Conference on Emerging Systems and Intelligent Computing, ESIC 2024

作者： Kumar, Sachin Mallick, Pradeep Kumar Vorfolomeyeva, Olga South Ural State University Big Data and Machine Learning Lab Chelyabinsk Russia Kiit Deemed to Be University School of Computer Engineering Odisha Bhubaneswar India West Kazakhstan State University Faculty of Arts and Culture Uralsk Kazakhstan

ISBN: (纸本)9798350349856

High-quality X-rays are now available to diagnose lung diseases with the help of radiologists. However, the diagnostic process is time consuming and depends on specialist availability in medical institutions. Patient information may include chest X-rays of varying quality, medical test results, doctors' notes and prescriptions, and medication details, among others. In this study, we present a model for classifying pulmonary diseases using multimodal data from patient clinical studies and radiographic images. Various methods were used to generate artificial samples for both images and tabular data on the laboratory study results during data preparation. We also proposed a method for establishing a correspondence between the generated modals. The late fusion architecture of the proposed multimodal model was implemented. We conducted experiments on pulmonary data-set with two modalities. Results shows that an increase in accuracy and other parameters were observed for multimodal data fusion using our model in comparison with image only modality and clinical data only modality. It strengthen the fact that multimodality provides more insight for the fusion model to learn and provide a more precise diagnosis than single modality. © 2024 IEEE.

关键词： Biological organs

来源：评论

学校读者我要写书评

暂无评论

Adversarial learning With Cost-Sensitive Classes

引用

IEEE TRANSACTIONS ON CYBERNETICS 2023年第8期53卷 4855-4866页

作者： Shen, Haojing Chen, Sihong Wang, Ran Wang, Xizhao Shenzhen Univ Big Data Inst Coll Comp Sci & Software Engn Guangdong Key Lab Intelligent Informat Proc Shenzhen 518060 Guangdong Peoples R China Shenzhen Univ Coll Math & Stat Shenzhen 518060 Peoples R China Shenzhen Univ Shenzhen Key Lab Adv Machine Learning & Applicat Shenzhen 518060 Peoples R China

It is necessary to improve the performance of some special classes or to particularly protect them from attacks in adversarial learning. This article proposes a framework combining cost-sensitive classification and adversarial learning together to train a model that can distinguish between the protected and unprotected classes, such that the protected classes are less vulnerable to adversarial examples. We find in this framework an interesting phenomenon during the training of deep neural networks, called the Min-Max property, that is, the absolute values of most parameters in the convolutional layer approach 0 while the absolute values of a few parameters are significantly larger, becoming bigger. Based on this Min-Max property which is formulated and analyzed in a view of random distribution, we further build a new defense model against adversarial examples for adversarial robustness improvement. An advantage of the built model is that it performs better than the standard one and can combine with adversarial training to achieve improved performance. It is experimentally confirmed that, regarding the average accuracy of all classes, our model is almost as same as the existing models when an attack does not occur and is better than the existing models when an attack occurs. Specifically, regarding the accuracy of protected classes, the proposed model is much better than the existing models when an attack occurs.

关键词： Training Robustness Costs Adversarial machine learning Perturbation methods Task analysis Neural networks Adversarial examples adversarial training attack and defense cost sensitive robustness

来源：评论

学校读者我要写书评

暂无评论

SmartHealth: An intelligent framework to secure IoMT service applications using machine learning

引用

ICT EXPRESS 2024年第2期10卷 425-430页

作者： Rani, Sita Kumar, Sachin Kataria, Aman Min, Hong Guru Nanak Dev Engn Coll Dept Comp Sci & Engn Ludhiana Punjab India South Ural State Univ Natl Res Univ Big Data & Machine Learning Lab Chelyabinsk Russia Amity Univ Amity Inst Def Technol Noida CampusSect 125 Noida Uttar Pradesh India Gachon Univ Sch Comp Seongnam South Korea

Due to the complex functioning of Smart Healthcare Systems (SHS), many security concerns have been raised in the past. It provisions the attackers to hamper the working of SHS in a variety of ways, e.g., injection of false data to replace vital signs, tampering of medical devices to prevent informing critical situations, etc. In this work, a novel ML-based framework, i.e., SmartHealth is proposed to secure IoMT devices in SHS. SmartHealth watches the vital signs gathered through different IoMT to analyze the change in various body activities to differentiate between normal activities and dangerous security attacks. The performance of the SmartHealth is also analyzed for three different dangerous attacks. During performance analysis, it has been observed that SmartHealth can identify wicked activities in IoMT 92% times accurately with an F1-score of 90%. (c) 2023 The Authors. Published by Elsevier B.V. on behalf of The Korean Institute of Communications and Information Sciences. This is an open access article under the CC BY license (http://***/licenses/by/4.0/).

关键词： Attacks Smart Healthcare Systems (SHS) Security Internet of Medical Things (ioMT) Service applications machine learning

来源：评论

学校读者我要写书评

暂无评论

Dynamic Co-Embedding Model for Temporal Attributed Networks

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2024年第3期35卷 3488-3502页

作者： Tang, Shaowei Meng, Zaiqiao Liang, Shangsong Sun Yat Sen Univ Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Guangdong Key Lab Big Data Anal & Proc Guangzhou 510006 Peoples R China Univ Glasgow Sch Comp Sci Glasgow G12 8QQ Lanark Scotland Mohamed Bin Zayed Univ Artificial Intelligence Dept Machine Learning Abu Dhabi U Arab Emirates

In this article, we study the problem of embedding temporal attributed networks, with the goal of which is to learn dynamic low-dimensional representations over time for temporal attributed networks. Existing temporal network embedding methods only learn the representations for nodes, which are unable to capture the dynamic affinities between nodes and attributes. Moreover, existing co-embedding methods that learn the static embeddings of both nodes and attributes cannot be naturally utilized to obtain their dynamic embeddings for temporal attributed networks. To address these issues, we propose the dynamic co-embedding model for temporal attributed networks (DCTANs) based on the dynamic stochastic state-space framework. Our model captures the dynamics of a temporal attributed network by modeling the abstract belief states representing the condition of the nodes and attributes of current time step, and predicting the transitions between temporal abstract states of two successive time steps. Our model is able to learn embeddings for both nodes and attributes based on their belief states at each time step of the temporal attributed network, while the state transition tendency for predicting the future network can be tracked and the affinities between nodes and attributes can be preserved. Experimental results on real-world networks demonstrate that our model achieves substantial performance gains in several static and dynamic graph mining applications compared with the state-of-the-art static and dynamic models.

关键词： Task analysis Predictive models Stochastic processes Analytical models Codes Network analyzers Correlation Dynamic co-embedding network embedding temporal attributed network variational autoencoder (VAE)

来源：评论

学校读者我要写书评

暂无评论

Relation-Aware Fine-Grained Reasoning Network for Textbook Question Answering

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2023年第1期34卷 15-27页

作者： Ma, Jie Liu, Jun Wang, Yaxian Li, Junjun Liu, Tongliang Xi An Jiao Tong Univ Sch Comp Sci & Technol Natl Engn Lab Big Data Analyt Xian 710049 Peoples R China Xi An Jiao Tong Univ Sch Comp Sci & Technol Shaanxi Prov Key Lab Satellite & Terr Network Tec Xian 710049 Peoples R China Univ Sydney Sch Comp Sci Fac Engn Trustworthy Machine Learning Lab Darlington NSW 2008 Australia

Textbook question answering (TQA) is a task that one should answer non-diagram and diagram questions accurately, given a large context which consists of abundant diagrams and essays. Although lots of studies have made significant progress in the natural image question answering (QA), they are not applicable to comprehending diagrams and reasoning over the long multimodal context. To address the above issues, we propose a relation-aware fine-grained reasoning (RAFR) network that performs fine-grained reasoning over the nodes of relation-based diagram graphs. Our method uses semantic dependencies and relative positions between nodes in the diagram to construct relation graphs and applies graph attention networks to learn diagram representations. To extract and reason over the multimodal knowledge, we first extract the text that is the most relevant to questions, options, and the instructional diagram which is the most relevant to question diagrams at the word-sentence level and the node-diagram level, respectively. Then, we apply instructional-diagram-guided attention and question-guided attention to reason over the node of question diagrams, respectively. The experimental results show that our proposed method achieves the best performance on the TQA dataset compared with baselines. We also conduct extensive ablation studies to comprehensively analyze the proposed method.

关键词： Semantics Cognition Knowledge discovery Task analysis Visualization Shape Frontal lobe Diagram understanding machine reading comprehension (MRC) multi-modality textbook question answering (TQA)

来源：评论

学校读者我要写书评

暂无评论

Point-Based Learnable Query Generator for Human–Object Interaction Detection

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2023年 32卷 6469-6484页

作者： Lin, Wang-Kai Zhang, Hong-Bo Fan, Zongwen Liu, Jing-Hua Yang, Li-Jie Lei, Qing Du, Jixiang Huaqiao Univ Dept Comp Sci & Technol Xiamen 361000 Peoples R China Huaqiao Univ Fujian Prov Univ Key Lab Comp Vis & Machine Learning Xiamen 361000 Peoples R China Huaqiao Univ Xiamen Key Lab Comp Vis & Pattern Recognit Xiamen 361000 Peoples R China Huaqiao Univ Fujian Key Lab Big Data Intelligence & Secur Xiamen 361000 Peoples R China

Transformer-based and interaction point-based methods have demonstrated promising performance and potential in human-object interaction detection. However, due to differences in structure and properties, direct integration of these two types of models is not feasible. Recent Transformer-based methods divide the decoder into two branches: an instance decoder for human-object pair detection and a classification decoder for interaction recognition. While the attention mechanism within the Transformer enhances the connection between localization and classification, this paper focuses on further improving HOI detection performance by increasing the intrinsic correlation between instance and action features. To address these challenges, this paper proposes a novel Transformer-based HOI Detection framework. In the proposed method, the decoder contains three parts: learnable query generator, instance decoder, and interaction classifier. The learnable query generator aims to build an effective query to guide the instance decoder and interaction classifier to learn more accurate instance and interaction features. These features are then applied to update the query generator for the next layer. Especially, inspired by the interaction point-based HOI and object detection methods, this paper introduces the prior bounding boxes, keypoints detection and spatial relation feature to build the novel learnable query generator. Finally, the proposed method is verified on HICO-DET and V-COCO datasets. The experimental results show that the proposed method has the better performance compared with the state-of-the-art methods.

关键词： Human-object interaction detection learnable query generator prior bounding boxes point detection spatial relation feature

来源：评论

学校读者我要写书评

暂无评论

SDN and application layer DDoS attacks detection in IoT devices by attention-based Bi-LSTM-CNN

引用

TRANSACTIONS ON EMERGING TELECOMMUNICATIONS TECHNOLOGIES 2023年第11期34卷

作者： Priyadarshini, Ishaani Mohanty, Pinaki Alkhayyat, Ahmed Sharma, Rohit Kumar, Sachin Univ Calif Berkeley Sch Informat Berkeley CA USA Purdue Univ Dept Comp Sci W Lafayette IN USA Islamic Univ Coll Tech Engn Najaf Iraq SRM Inst Sci & Technol Dept Elect & Commun Engn Delhi NCR Campus Ghaziabad Uttar Pradesh India South Ural State Univ Big Data & Machine Learning Lab Chelyabinsk Russia

The Internet of Things (IoT) is connecting more devices every day. Security is critical to ensure that the devices operate in a trusted environment. The lack of proper IoT security encourages cybercriminals to target many smart devices across the network and gain sensitive information. Distributed Denial of Service (DDoS) attacks are common in the IoT infrastructure and involve hijacking IoT devices to consume resources and interrupt services. This may specifically vandalize the application running the service that the end users are trying to access (application layer DDoS attacks) or flood the network bandwidth leading to network failure (software defined network DDoS attacks). This article proposes a hybrid attention-based bidirectional long short term memory (LSTM) with convolutional neural networks (CNN) to identify DDoS attacks in the application layer and SDN. We deploy several other machine learning models like logistic regression, decision trees, random forests, support vector machines, K-nearest neighbors, extreme gradient boosting, artificial neural networks, CNN, LSTM, CNN-LSTM to evaluate the performance of our proposed model. The evaluation metrics considered for the study are accuracy, precision, recall, and F-1 score. The experimental analysis on multiple datasets exhibits that the proposed model performs the classification efficiently with an accuracy of 99.74% and 99.98%.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Diffusion-Based Method with TTS Guidance for Foreign Accent Conversion

Diffusion-Based Method with TTS Guidance for Foreign Accent ...

引用

International Symposium on Chinese Spoken Language Processing

作者： Qibing Bai Shuai Wang Zhijun Liu Mingyang Zhang Wei Rao Yannan Wang Haizhou Li School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen Tencent Ethereal Audio Lab Tencent Machine Learning Lab (MLL) University of Bremen

ISBN: (数字)9798331516826

ISBN: (纸本)9798331516833

Accent conversion (AC) aims to alter the accent of spoken language while preserving the original content and speaker characteristics. While any accent can be selected as a target, foreign accent conversion (FAC) that focuses on L2 speakers is particularly noteworthy due to its wide-ranging applications. Compared to general voice conversion tasks, which focus on speaker conversion, research related to accent conversion is relatively scarce, and the audio quality is often limited. In this article, we introduce a diffusion decoder into the conventional TTS-guided accent conversion framework and propose a phoneme-level acoustic-linguistic alignment strategy. Subjective evaluations on the Chinese-accent source speech confirm that the proposed method outperforms the baseline in terms of speech naturalness, accentedness, and speaker similarity 1 1 Audio samples are available at ***..

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

LmTag: functional-enrichment and imputation-aware tag SNP selection for population-specific genotyping arrays

引用

BRIEFINGS IN BIOINFORMATICS 2022年第4期23卷 bbac252-bbac252页

作者： Dat Thanh Nguyen Quan Hoang Nguyen Nguyen Thuy Duong Vo, Nam S. Vingrp Big Data Inst Ctr Biomed Informat Hanoi Vietnam Univ Queensland UQ Inst Mol Biosci IMB Genom & Machine Learning Lab Brisbane Qld Australia Vin Univ Coll Engn & Comp Sci Vinhome Ocean Pk Hanoi Vietnam

Despite the rapid development of sequencing technology, single-nucleotide polymorphism (SNP) arrays are still the most cost-effective genotyping solutions for large-scale genomic research and applications. Recent years have witnessed the rapid development of numerous genotyping platforms of different sizes and designs, but population-specific platforms are still lacking, especially for those in developing countries. SNP arrays designed for these countries should be cost-effective (small size), yet incorporate key information needed to associate genotypes with traits. A key design principle for most current platforms is to improve genome-wide imputation so that more SNPs not included in the array (imputed SNPs) can be predicted. However, current tag SNP selection methods mostly focus on imputation accuracy and coverage, but not the functional content of the array. It is those functional SNPs that are most likely associated with traits. Here, we propose LmTag, a novel method for tag SNP selection that not only improves imputation performance but also prioritizes highly functional SNP markers. We apply LmTag on a wide range of populations using both public and in-house whole-genome sequencing databases. Our results show that LmTag improved both functional marker prioritization and genome-wide imputation accuracy compared to existing methods. This novel approach could contribute to the next generation genotyping arrays that provide excellent imputation capability as well as facilitate array-based functional genetic studies. Such arrays are particularly suitable for under-represented populations in developing countries or non-model species, where little genomics data are available while investment in genome sequencing or high-density SNP arrays is limited. LmTag is available at: https://***/datngu/LmTag.

关键词： tag SNP selection SNP array design linkage disequilibrium beam search

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

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

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：