检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Liu, Zhiqiang Li, Yuhong Huang, Chengkai Liu, Yanxia School of software Engineering South China University of Technology China School of Computer Science and Engineering The University of New South Wales Australia

Fine-tuning is an effective technique to enhance network performance in scenarios with limited labeled data. To achieve this, recent methods exploit the knowledge mined in the source model (e.g., feature maps) to construct an extra regularization signal (RS), collaboratively supervising the target model along with target labels. However, these RSs are generated independently from the target information or are generated from the rough assistance of the target information, resulting in biased supervision different from the target task. In this paper, we propose a Conditional Online Knowledge Transfer (COKT) framework that finely utilizes the target information to construct robust and target-related RS. Specifically, we train a target-dominant RS branch that online supervises the target model in a knowledge distillation manner. The target information dominates the RS branch from three aspects: sample-wise conditional attention, residual feature fusion, and target task loss. With such a target-oriented framework, we can effectively exploit target-related prior knowledge of the source model. Extensive experiments demonstrate that COKT significantly outperforms the fine-tuning baselines, especially for dissimilar target tasks and small datasets. Moreover, different from most of the fine-tuning methods that are restricted to the vanilla fine-tuning scenario, COKT can be easily extended to cross-model and multimodel fine-tuning scenarios. © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Beyond Subspace Isolation: Many-to-Many Transformer for Light Field Image Super-resolution

arXiv

引用

arXiv 2024年

作者： Hu, Zeke Zexi Chen, Xiaoming Chung, Vera Yuk Ying Shen, Yiran School of Computer Science University of Sydney DarlingtonNSW2008 Australia School of Computer Science and Engineering Beijing Technology and Business University Beijing102488 China School of Software Shandong University Jinan250100 China

The effective extraction of spatial-angular features plays a crucial role in light field image super-resolution (LFSR) tasks, and the introduction of convolution and Transformers leads to significant improvement in this area. Nevertheless, due to the large 4D data volume of light field images, many existing methods opted to decompose the data into a number of lower-dimensional subspaces and perform Transformers in each subspace individually. As a side effect, these methods inadvertently restrict the self-attention mechanisms to a One-to-One scheme accessing only a limited subset of LF data, explicitly preventing comprehensive optimization on all spatial and angular cues. In this paper, we identify this limitation as subspace isolation and introduce a novel Many-to-Many Transformer (M2MT) to address it. M2MT aggregates angular information in the spatial subspace before performing the self-attention mechanism. It enables complete access to all information across all sub-aperture images (SAIs) in a light field image. Consequently, M2MT is enabled to comprehensively capture long-range correlation dependencies. With M2MT as the foundational component, we develop a simple yet effective M2MT network for LFSR. Our experimental results demonstrate that M2MT achieves state-of-the-art performance across various public datasets, and it offers a favorable balance between model performance and efficiency, yielding higher-quality LFSR results with substantially lower demand for memory and computation. We further conduct in-depth analysis using local attribution maps (LAM) to obtain visual interpretability, and the results validate that M2MT is empowered with a truly non-local context in both spatial and angular subspaces to mitigate subspace isolation and acquire effective spatial-angular representation. Copyright © 2024, The Authors. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Mobile Based Solution to Weight Loss Planning for Children (with Obesity) in Sri Lanka 3

Mobile Based Solution to Weight Loss Planning for Children (...

引用

3rd International Conference on Advancements in Computing, ICAC 2021

作者： Rajapakse, R.M.M.P.K. Mudalige, J.M.A.I. Perera, L.A.D.Y.S. Warakagoda, R.N.A.M.S.C.B. Siriwardana, Samanthi Sri Lanka Institute of Information Technology Department of Computer Science and Software Engineering Malabe Sri Lanka

ISBN: (纸本)9781665408622

Obesity is a condition where there is excess fat in the body, and it is one of the world's most extreme and dangerous dietary diseases. Genetic factors, lack of physical activity, unhealthy eating patterns, or a combination of these factors are the most common causes of obesity. This is important because it influences every part of a child's life. More, in particular, this disorder leads to poor health and negative social standing with perceptions. Nowadays, children are paying keen interest in technology and related devices. Therefore, in this research, we are planning to give a mobile-based solution with a smart band that is used to monitor the child. In this solution, we are mainly focusing on Sri Lankan children with obesity who are aged between 5-10. In our solution, there are four main sections which are, monitoring child activities, recognizing the activities, and getting relevant data, then based on those data and previous activity completion levels, this solution will suggest activities for losing weight, provide specific diet plans for each child considering the health conditions and predict the probability of having main obesity-related diseases by using previously gathered data. © 2021 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

ETHACA - A Container-Based Cyber Range for Research and Education 29

ETHACA - A Container-Based Cyber Range for Research and Educ...

引用

29th IEEE International Workshop on computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2024

作者： Chouliaras, Nestoras Kantzavelou, Ioanna Pantziou, Grammati Maglaras, Leandros Chouliara-Sidera, Panagiota Adamakos, Antonis Chalatsakos, Panagiotis University of West Attica Department of Informatics and Computer Engineering Athens Greece School of Computing Edinburgh Napier University EdinburghEH10 5DT United Kingdom University of Thessaly Department of Computer Science and Biomedical Infromatics Lamia Greece Senior Software Engineer at *** Athens Greece Open University of Cyprus Faculty of Pure and Applied Sciences Nicosia Cyprus

ISBN: (纸本)9798350377644

The advent of Industry 4.0 and the Industrial Internet of Things (IIoT) has revolutionized manufacturing and production systems, enhancing efficiency and communication through the integration of advanced technologies such as IoT devices and AI. However, this increased digitization has also led to heightened cybersecurity risks, with a notable rise in the frequency and complexity of cyber-attacks. Cyber Ranges emerges as a critical solution in addressing these challenges, offering a controlled environment for simulating cyber-attack scenarios, training cybersecurity teams, and developing robust defense strategies. The escalating demand for cybersecurity professionals has highlighted the crucial need for effective training solutions. This article presents an open-source Cyber Range constructed on container-based architecture, meticulously engineered for efficient operation with minimal resource consumption. In response to the scarcity of cybersecurity experts, this platform prioritizes extensive scalability and robust isolation between exercises, ensuring secure learning environments, and also integrates a complex developed exercise series aimed at strengthening cybersecurity learning experiences. These exercises encompass a variety of scenarios and challenges, catering to professionals at different skill levels. Each exercise replicates real-world cyber threats, offering practical, hands-on learning opportunities. Additionally, a comprehensive questionnaire accompanies these exercises, designed to gather feedback on participants' experiences. © 2024 IEEE.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

FLER-Net: Federated Learning for Emotion Recognition Using EEG Signals and Dense Neural Networks

SSRN

引用

SSRN 2023年

作者： Anwar, Mohd Ayaan Agrawal, Manan Sethia, Divyashikha Department of Computer Science and Engineering Delhi Technological University Delhi India Department of Software Engineering Delhi Technological University Delhi India

Emotion recognition techniques in current literature tend to follow a centralized training process that puts user privacy at risk. Federated Learning (FL) can fill these gaps by utilizing the principles of iterative decentralized learning, data-sharding, and weighted averaging of gradients from individual training of local models. This paper presents a new emotion recognition framework called FLER-Net (Federated Learning for Emotion Recognition using EEG Signals and Dense Neural Networks). It follows a privacy-preserving method of training a server-hosted model without accessing users' private data. FLER-Net has been trained and evaluated on the DEAP dataset of Electroencephalogram (EEG) signals. FLER-Net achieves accuracy scores of 89.30%, 86.85%, and 88.16% for the Valence, Arousal, and Dominance labels. Compared to previous work on emotion recognition with privacy preservation, FLER-Net achieves higher accuracy by utilizing a more extensive feature space and model architecture optimized for federated learning. © 2023, The Authors. All rights reserved.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

PATCH: Empowering Large Language Model with Programmer-Intent Guidance and Collaborative-Behavior Simulation for Automatic Bug Fixing

arXiv

引用

arXiv 2025年

作者： Zhang, Yuwei Jin, Zhi Xing, Ying Li, Ge Liu, Fang Zhu, Jiaxin Dou, Wensheng Wei, Jun Affiliated with Nanjing Institute of Software Technology University of Chinese Academy of Sciences Nanjing China Key Laboratory of System Software Chinese Academy of Sciences Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences Beijing China Key Laboratory of High Confidence Software Technologies Peking University Ministry of Education School of Computer Science Peking University Beijing China School of Computer Science Wuhan University Wuhan China School of Intelligent Engineering and Automation Beijing University of Posts and Telecommunications Beijing China State Key Laboratory of Complex & Critical Software Environment School of Computer Science and Engineering Beihang University Beijing China

Bug fixing holds significant importance in software development and maintenance. Recent research has made substantial strides in exploring the potential of large language models (LLMs) for automatically resolving software bugs. However, a noticeable gap in existing approaches lies in the oversight of collaborative facets intrinsic to bug resolution, treating the process as a single-stage endeavor. Moreover, most approaches solely take the buggy code snippet as input for LLMs during the patch generation stage. To mitigate the aforementioned limitations, we introduce a novel stage-wise framework named PATCH. Specifically, we first augment the buggy code snippet with corresponding dependence context and intent information to better guide LLMs in generating the correct candidate patches. Additionally, by taking inspiration from bug management practices, we decompose the bug-fixing task into four distinct stages: bug reporting, bug diagnosis, patch generation, and patch verification. These stages are performed interactively by LLMs, aiming to simulate the collaborative behavior of programmers during the resolution of software bugs. By harnessing these collective contributions, PATCH effectively enhances the bug-fixing capability of LLMs. We implement PATCH by employing the powerful dialogue-based LLM ChatGPT. Our evaluation on the widely used bug-fixing benchmark BFP demonstrates that PATCH has achieved better performance than state-of-the-art LLMs. Copyright © 2025, The Authors. All rights reserved.

关键词： computer software maintenance

来源：评论

学校读者我要写书评

暂无评论

An empirical study of deep learning-based feature extractor models for imbalanced image classification

引用

Advances in Computational Intelligence 2023年第6期3卷 1-16页

作者： Khan, Ammara Rasheed, Muhammad Tahir Khan, Hufsa Department of Botany University of the Punjab Lahore Pakistan College of Computer Science and Software Engineering Shenzhen University Shenzhen China College of Electronics and Information Engineering Shenzhen university Shenzhen China

Deep learning has played an important role in many real-life applications, especially in image classification. It is often found that some domain data are highly skewed, i.e., most of the data belongs to a handful of majority classes, and the minority classes only contain small amounts of information. It is important to acknowledge that skewed class distribution poses a significant challenge to machine learning algorithms. Due to which in case of imbalanced data distribution, the majority of machine and deep learning algorithms are not effective or may fail when it is highly imbalanced. In this study, a comprehensive analysis in case of imbalanced dataset is performed by considering deep learning based well known models. In particular, the best feature extractor model is identified and the current trend of latest feature extraction model is investigated. Moreover, to determine the global scientific research on the image classification of imbalanced mushroom dataset, a bibliometric analysis is conducted from 1991 to 2022. In summary, our findings may offer researchers a quick benchmarking reference and alternative approach to assessing trends in imbalanced data distributions in image classification research.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Short-Time Traffic Flow Prediction Based on a Combined Model of EMD and LSTM

Short-Time Traffic Flow Prediction Based on a Combined Model...

引用

Artificial Intelligence, Networking and Information Technology (AINIT), International Seminar on

作者： Qihan Zhao Lidu Lou Bo Ouyang School of Computer Science and Engineering Central South University Changsha China School of Software Engineering South China University Hengyang China

In traffic management, accurate forecasting of short-term traffic patterns is of utmost importance to achieve optimal performance and efficiency of road networks. This research proposes a prediction technique for short-term traffic flow, which utilizes empirical modal decomposition (EMD) and long short-term memory neural networks (LSTM). Firstly, the traffic flow sequence is decomposed into a series of relatively stable subseries using EMD, minimizing the impact of various trend data interactions. Secondly, to improve model training efficiency, normalization is applied separately to each subseries. Subsequently, an LSTM-based time-series prediction model is built for each subseries, which enhances the model's predictive accuracy. Finally, the forecasted values of short-term traffic flow are obtained by aggregating the prediction outcomes of each subseries. The simulation results demonstrate that the proposed method more accurately predicts the traffic flow change trend and achieves higher stability than conventional prediction techniques.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FedLE: Federated Learning Client Selection with Lifespan Extension for Edge IoT Networks

arXiv

引用

arXiv 2023年

作者： Wu, Jiajun Drew, Steve Zhou, Jiayu Electrical and Software Engineering University of Calgary Calgary Canada Computer Science and Engineering Michigan State University East Lansing United States

Federated learning (FL) is a distributed and privacy-preserving learning framework for predictive modeling with massive data generated at the edge by Internet of Things (IoT) devices. One major challenge preventing the wide adoption of FL in IoT is the pervasive power supply constraints of IoT devices due to the intensive energy consumption of battery-powered clients for local training and model updates. Low battery levels of clients eventually lead to their early dropouts from edge networks, loss of training data jeopardizing the performance of FL, and their availability to perform other designated tasks. In this paper, we propose FedLE, an energy-efficient client selection framework that enables lifespan extension of edge IoT networks. In FedLE, the clients first run for a minimum epoch to generate their local model update. The models are partially uploaded to the server for calculating similarities between each pair of clients. Clustering is performed against these client pairs to identify those with similar model distributions. In each round, low-powered clients have a lower probability of being selected, delaying the draining of their batteries. Empirical studies show that FedLE outperforms baselines on benchmark datasets and lasts more training rounds than FedAvg with battery power constraints. © 2023, CC BY.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Sliced Maximal Information Coefficient: A Training-Free Approach for Image Quality Assessment Enhancement

Sliced Maximal Information Coefficient: A Training-Free Appr...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Kang Xiao Xu Wang Yulin He Baoliang Chen Xuelin Shen Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) Shenzhen China College of Computer Science and Software Engineering Shenzhen University Shenzhen China School Of Computer Science South China Normal University Guangzhou China

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Full-reference image quality assessment (FR-IQA) models generally operate by measuring the visual differences between a degraded image and its reference. However, existing FR-IQA models including both the classical ones (e.g., PSNR and SSIM) and deep-learning based measures (e.g., LPIPS and DISTS) still exhibit limitations in capturing the full perception characteristics of the human visual system (HVS). In this paper, instead of designing a new FR-IQA measure, we aim to explore a generalized human visual attention estimation strategy to mimic the process of human quality rating and enhance existing IQA models. In particular, we model human attention generation by measuring the statistical dependency between the degraded image and the reference image. The dependency is captured in a training-free manner by our proposed sliced maximal information coefficient and exhibits surprising generalization in different IQA measures. Experimental results verify the performance of existing IQA models can be consistently improved when our attention module is incorporated. The source code is available at https://***/KANGX99/SMIC.

关键词： Image quality Visualization Microwave integrated circuits Atmospheric measurements Source coding Estimation Visual systems Particle measurements Distortion

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：