检索结果-内蒙古大学图书馆

Network slimming using Lp(p<1) regularization

Pattern Recognition 2025年 167卷

作者： Ji, Fengrui Chen, Xinwang Chu, Renxin Liu, Baolin University of Science and Technology Beijing School of Computer and Communication Engineering Beijing 100083 China Harvard Medical School BostonMA02115 United States

Convolutional neural networks (CNNs) require significant computational resources, making them challenging to deploy on battery-powered devices. Channel pruning provides a practical solution by reducing model size to facilitate efficient deployment. Among various pruning methods, L0 regularization is an effective approach for identifying sparse sub-networks. However, solving it directly is NP-hard. To overcome this challenge, we approximate the L0 norm using Lp regularization with pp regularization term introduces a non-convex subproblem, we linearize it to transform the subproblem into a convex optimization problem. Nevertheless, when the model parameter approaches zero, the Lp norm grows excessively, leading to no convergence in ADMM. To address this, we introduce a synaptic inhibition mechanism to suppress the excessive growth of the Lp regularization term, ensuring ADMM algorithm convergence. Experiments on CIFAR, ImageNet, and GLUE benchmark dataset demonstrate that our method achieves state-of-the-art performance in channel pruning. © 2025 Elsevier Ltd

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

iCTS: Iterative and Hierarchical Clock Tree Synthesis With Skew-Latency-Load Tree

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Li, Weiguo Huang, Zhipeng Yu, Bei Zhu, Wenxing Chen, Jian He, Zhixue Li, Xingquan South China University of Technology School of Software Engineering Guangzhou China Pengcheng Laboratory Shenzhen China Beijing Institute of Open Source Chip Beijing China Chinese University of Hong Kong Department of Computer Science and Engineering Hong Kong Hong Kong Fuzhou University Center for Discrete Mathematics and Theoretical Computer Science Fuzhou China

The advancement of modern clock tree synthesis (CTS) encounters a bottleneck, primarily due to the difficulty in achieving multi-objective co-optimization among complex design processes. To concurrently optimize skew, latency, and load capacitance, we propose an iterative and hierarchical CTS framework, which is composed of clustering, topology generation and routing, buffering, and optimization. First, we introduce a capacitance-based metric to achieve adaptive balanced clustering and optimize the cluster results through simulated annealing. Second, to construct a clock tree with lower latency, load capacitance, and skew, we introduce the skew-latency-load tree (SLLT), which combines the advantages of bound skew tree and Steiner shallow-light tree, and we propose an effective SLLT construction algorithm. Third, to further optimize CTS result by buffering, we introduce the critical wirelength evaluation (CWE) to evaluate the capability of each buffer, and propose the insertion delay estimation (IDE) to reduce the evaluation bias during buffering, then design the iterative skew convergence algorithm (ISCA) to achieve complete convergence of skew. We validate our solution using 28nm process technology. Compared to our method, the commercial tool increases skew, latency, and clock capacitance by 39.5%, 13.0%, and 18.5%, respectively, while the OpenROAD by 101.6%, 50.7%, and 25.5%, respectively. © 1982-2012 IEEE.

关键词： Simulated annealing

来源：评论

学校读者我要写书评

暂无评论

DP-Adapter: Dual-Pathway Adapter for Boosting Fidelity and Text Consistency in Customizable Human Image Generation

arXiv

引用

arXiv 2025年

作者： Wang, Ye Xie, Xuping Wang, Lanjun Yi, Zili Ma, Rui School of Artificial Intelligence Jilin University China College of Computer Science and Technology Jilin University China School of New Media and Communication Tianjin University China School of Intelligence Science and Technology Nanjing University China Engineering Research Center of Knowledge-Driven Human-Machine Intelligence MOE China

With the growing popularity of personalized human content creation and sharing, there is a rising demand for advanced techniques in customized human image generation. However, current methods struggle to simultaneously maintain the fidelity of human identity and ensure the consistency of textual prompts, often resulting in suboptimal outcomes. This shortcoming is primarily due to the lack of effective constraints during the simultaneous integration of visual and textual prompts, leading to unhealthy mutual interference that compromises the full expression of both types of input. Building on prior research that suggests visual and textual conditions influence different regions of an image in distinct ways, we introduce a novel Dual-Pathway Adapter (DP-Adapter) to enhance both high-fidelity identity preservation and textual consistency in personalized human image generation. Our approach begins by decoupling the target human image into visually sensitive and text-sensitive regions. For visually sensitive regions, DP-Adapter employs an Identity-Enhancing Adapter (IEA) to preserve detailed identity features. For text-sensitive regions, we introduce a Textual-Consistency Adapter (TCA) to minimize visual interference and ensure the consistency of textual semantics. To seamlessly integrate these pathways, we develop a Fine-Grained Feature-Level Blending (FFB) module that efficiently combines hierarchical semantic features from both pathways, resulting in more natural and coherent synthesis outcomes. Additionally, DP-Adapter supports various innovative applications, including controllable headshot-to-full-body portrait generation, age editing, old-photo to reality, and expression editing. Extensive experiments demonstrate that DP-Adapter outperforms state-of-the-art methods in both visual fidelity and text consistency, highlighting its effectiveness and versatility in the field of human image generation. © 2025, CC BY.

关键词： Photointerpretation

来源：评论

学校读者我要写书评

暂无评论

Multi-Label Movie Genre Classification with Attention Mechanism on Movie Plots

引用

computers, Materials and Continua 2025年第3期83卷 5595-5622页

作者： Faheem Shaukat Naveed Ejaz Rashid Kamal Tamim Alkhalifah Sheraz Aslam Mu Mu Department of Computing and Technology Iqra University H-9 Campus Islamabad 04436 Pakistan School of Computing Queens University Kingston ON K7L2N8 Canada School of Computing Ulster University Belfast BT15 1ED Northern Ireland UK Faculty of Arts Science and Technology University of Northampton Waterside Campus Northampton Northamptonshire NN1 5PH UK Department of Computer Engineering College of Computer Qassim University Buraydah Saudi Arabia Department of Computer Science CTL Eurocollege Limassol 3077 Cyprus Department of Electrical Engineering Computer Engineering and Informatics Cyprus University of Technology Limassol 3036 Cyprus

Automated and accurate movie genre classification is crucial for content organization, recommendation systems, and audience targeting in the film industry. Although most existing approaches focus on audiovisual features such as trailers and posters, the text-based classification remains underexplored despite its accessibility and semantic richness. This paper introduces the Genre Attention Model (GAM), a deep learning architecture that integrates transformer models with a hierarchical attention mechanism to extract and leverage contextual information from movie plots for multi-label genre classification. In order to assess its effectiveness, we assess multiple transformer-based models, including Bidirectional Encoder Representations from Transformers (BERT), A Lite BERT (ALBERT), Distilled BERT (DistilBERT), Robustly Optimized BERT Pretraining Approach (RoBERTa), Efficiently Learning an Encoder that Classifies Token Replacements Accurately (ELECTRA), eXtreme Learning Network (XLNet) and Decoding-enhanced BERT with Disentangled Attention (DeBERTa). Experimental results demonstrate the superior performance of DeBERTa-based GAM, which employs a two-tier hierarchical attention mechanism: word-level attention highlights key terms, while sentence-level attention captures critical narrative segments, ensuring a refined and interpretable representation of movie plots. Evaluated on three benchmark datasets Trailers12K, Large Movie Trailer Dataset-9 (LMTD-9), and MovieLens37K. GAM achieves micro-average precision scores of 83.63%, 83.32%, and 83.34%, respectively, surpassing state-of-the-art models. Additionally, GAM is computationally efficient, requiring just 6.10 Giga Floating Point Operations Per Second (GFLOPS), making it a scalable and cost-effective solution. These results highlight the growing potential of text-based deep learning models in genre classification and GAM’s effectiveness in improving predictive accuracy while maintaining computational efficiency. With it

关键词： Multi-label classification artificial intelligence movie genre classification hierarchical attention mechanisms natural language processing content recommendation text-based genre classification explainable AI (Artificial Intelligence) transformer models BERT

来源：评论

学校读者我要写书评

暂无评论

Improving the Classification of Fruit Images: Denoising Techniques for Better Dataset Quality

Improving the Classification of Fruit Images: Denoising Tech...

引用

International Conference on Inventive Computation Technologies (ICICT)

作者： Rohini Mano Desai Varsha Yogesh Bhole Manoj M. Deshpande Department of Information Technology & Data Science Vidyalanakar School of Information Technology Wadala Mumbai Department of Computer Engineering A. C. Patil College of Engineering Navi Mumbai India Department of Information Technology A. C. Patil College of Engineering Navi Mumbai India

ISBN: (数字)9798331512248

ISBN: (纸本)9798331512255

The performance of machine learning models in the computer vision for the task of picture classification has greatly affected the quality of image datasets. Since noise from images can reduce the accuracy of model, noise reduction is a key preprocessing step. To clean the noisier images from a mixed fruit collection, this work investigates the application of enhanced filtering techniques, including PDTMF, CWF, GBF, and SIGF. This study aims to determine how different filters enhance image quality and how well they preserve the essential features for accurate fruit identification. The ability of each method is evaluated by applying standard image quality measures such as the Structural Similarity Index (SSIM), the Peak Signal-to-Noise Ratio (PSNR), and visual inspection. It has been found that though GBF and SIGF are good in removing some types of noise, i.e., Gaussian and salt and pepper noise respectively, PDMTF and CWF are the best in removing noise with least information loss about the image. So, this paper contributes to developing more reliable picture preprocessing techniques for fruit classification systems to enhance the accuracy and reliability of the automatic fruit recognition systems.

关键词： Image quality Visualization Accuracy PSNR Filters Wiener filters Computational modeling Noise Reliability Standards

来源：评论

学校读者我要写书评

暂无评论

Leveraging Deep Learning for COVID-19 Detection: A Comparative Study Using ResNet Models on X-Ray Images

Leveraging Deep Learning for COVID-19 Detection: A Comparati...

引用

Interdisciplinary Approaches in technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Disha Tuteja Anurag Sinha Deepak Chandra Uprety Dyuti Banerjee Ayushi Dewan Chetna Sharma Department of ECE College Guru Tegh Bahadur Institute of Technology Delhi India Department of Computer Science IGNOU New Delhi India Department of Computer Science and Engineering (CS) School of Computer Science in Emerging Technologies Noida Institute of Engineering and Technology Greater Noida India Department of Artificial Intelligence & Data Science (AI&DS) Koneru Lakshmaiah Education Foundation Green Fields Andhra Pradesh India Rawal Institutions Faridabad India Chitkara University Institute of Engineering and Technology Chitkara University Punjab India

ISBN: (数字)9798331521691

ISBN: (纸本)9798331521707

The COVID-19 pandemic has rapidly spread across the globe, infecting millions of people, with no signs of slowing down. This continuous surge in cases presents a major challenge for healthcare systems worldwide. To combat the exponential rise in infections, medical researchers have primarily relied on RT-PCR testing to distinguish between COVID-19 positive and negative cases. However, this approach faces significant hurdles, particularly in low-income countries where these tests are often too expensive and inaccessible in rural areas. Another critical challenge is the shortage of skilled healthcare professionals, especially doctors and radiologists, whose expertise is in high demand for diagnosing and treating COVID-19 patients. As a result, healthcare resources have been stretched to their limits. To address these issues, innovative solutions are essential, and artificial intelligence (AI) and machine learning (ML) offer promising possibilities. In medical diagnostics, AI, ML, and deep learning (DL) have demonstrated remarkable accuracy. This study explores the application of DL models, specifically ResNet-34 and ResNet-50, for COVID-19 detection using X-ray images. The results indicate an impressive accuracy of 96.4% with ResNet-34. To comprehensively evaluate the performance of these models, the study employs various quantitative metrics, including Matthews Correlation Coefficient (MCC), F1 Score, Precision, and Recall. By contributing to both the effective detection of COVID-19 and addressing the shortage of healthcare professionals, this research provides a hopeful step forward in the ongoing fight against the pandemic.

关键词： COVID-19 Deep learning Measurement Technological innovation Accuracy Pandemics Medical services Public healthcare X-ray imaging Testing

来源：评论

学校读者我要写书评

暂无评论

STMAE: Spatial Temporal Masked Auto-Encoder for Traffic Forecasting 27th

STMAE: Spatial Temporal Masked Auto-Encoder for Traffic For...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Wu, Xing Cai, Chengyou Wang, Xiaoxiao Wang, Jianjia Yao, Junfeng Qian, Quan Song, Jun School of Computer Engineering and Science Shanghai University Shanghai China Shanghai Institute for Advanced Communication and Data Science Shanghai University Shanghai China Shanghai Shentong Metro Co. Ltd. Shanghai China School of AI and Advanced Computing XJTLU Entrepreneur College Xi’an Jiaotong-Liverpool University Suzhou China CSSC Seago System Technology Co. Ltd. Shanghai China Department of Geography Hong Kong Baptist University Hong Kong

ISBN: (纸本)9783031781681

The advancement of intelligent transportation systems underscores the importance of data-driven approaches in traffic forecasting, which plays a crucial role in tasks such as traffic signal control and route guidance, among others. However, the inherent uncertainty stemming from regional traffic dynamics, coupled with intricate spatio-temporal correlations, poses formidable challenges to accurate traffic prediction. Moreover, the complexities inherent in sequence forecasting across varying scales further exacerbate the accuracy dilemma. Recognizing the need for integrating information across spatial and temporal dimensions to enhance forecasting precision, a novel solution termed Spatial Temporal Masked Autoencoder (STMAE) is introduced. The STMAE framework addresses these challenges through a two-stage learning process. In the pre-training phase, an autoencoder architecture is employed to extract spatio-temporal features from the data. In the fine-tuning phase, the pre-trained encoder of the STMAE model undergoes further refinement to specifically target traffic forecasting tasks. Extensive evaluations validate the effectiveness of the proposed STMAE model. Notably, STMAE demonstrates competitive performance, achieving 3.32 Vehs MAE for long-term (60 min) traffic forecasting while operating within a reduced computational budget. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Text-to-Decision Agent: Learning Generalist Policies from Natural Language Supervision

arXiv

引用

arXiv 2025年

作者： Zhang, Shilin Hu, Zican Wu, Wenhao Xie, Xinyi Tang, Jianxiang Chen, Chunlin Dong, Daoyi Cheng, Yu Sun, Zhenhong Wang, Zhi Department of Control Science and Intelligent Engineering Nanjing University China School of Engineering Australian National University Australia Australian Artificial Intelligence Institute University of Technology Sydney Australia Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong

RL systems usually tackle generalization by inferring task beliefs from high-quality samples or warmup explorations. The restricted form limits their generality and usability since these supervision signals are expensive and even infeasible to acquire in advance for unseen tasks. Learning directly from the raw text about decision tasks is a promising alternative to leverage a much broader source of supervision. In the paper, we propose Text-to-Decision Agent (T2DA), a simple and scalable framework that supervises generalist policy learning with natural language. We first introduce a generalized world model to encode multi-task decision data into a dynamics-aware embedding space. Then, inspired by CLIP, we predict which textual description goes with which decision embedding, effectively bridging their semantic gap via contrastive language-decision pre-training and aligning the text embeddings to comprehend the environment dynamics. After training the text-conditioned generalist policy, the agent can directly realize zero-shot text-to-decision generation in response to language instructions. Comprehensive experiments on MuJoCo and Meta-World benchmarks show that T2DA facilitates high-capacity zero-shot generalization and outperforms various types of baselines. Copyright © 2025, The Authors. All rights reserved.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Optimized Signcryption Mechanisms with SIMON and Ring Signatures for Healthcare Data Security

Optimized Signcryption Mechanisms with SIMON and Ring Signat...

引用

International Conference on ent Computing and Control Systems (ICICCS)

作者： Desetti Srinuvasa Rao Swathi Balija Rohit Kanauzia Ankit Garg K. Chandramouli Balajee Maram Department of CSE GMRIT Rajam Andhra Pradesh School of Computer Science and Artificial Intelligence SR University Warangal Department of Computer Science and Engineering Chandigarh University AIT-CSE UCRD Chandigarh University Mohali Punjab India Aditya Institute of Technology and Management(AITAM) Tekkali AP School of Computer Science and Artificial Intelligence SR University Warangal Telangana

ISBN: (数字)9798331512088

ISBN: (纸本)9798331512095

With the rapid digitization in the health domain, needs arise to protect and safeguard sensitive information about patients in a secure, efficient, and prompt manner. This paper proposes an optimization mechanism for the protection of health data using techniques that apply ring signature-integrated lightweight block cipher SIMON. The SIMON encryption algorithm has wide-ranging speedy encryption that can be up to 25% faster than traditional block ciphers with a 30% saving in computational overheads and appropriate for resource-constrained environments. Ring signatures ensure there is anonymity of data yet authenticated in transmitting data. Therefore, this kind of proposed scheme will have higher integrity and confidentiality of data. The medical dataset was tested against the design; it even passed 98.5% attacks from cryptography. Encryption latency is still below 2 milliseconds. All the results ensure usability in modern systems of healthcare.

关键词： Ciphers Data privacy Quantum computing Scalability Medical services Encryption Usability Protection SIMO Optimization

来源：评论

学校读者我要写书评

暂无评论

Wireless Networked Adaptive Active Noise Control With Online Secondary Path Modeling

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 1078-1087页

作者： Chuang Shi Hao Yu Feiyu Du Mingzhe Li Le Zhang Institute of Sound and Vibration Research University of Southampton Southampton U.K. School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China Department of Computer Science City University of Hong Kong Hong Kong SAR China

Active noise control (ANC) has been developed over the years to mitigate unwanted noise. However, the fixed placement of electroacoustic components often limits ANC performance. To address this constraint, wireless networked ANC (WNANC) has been proposed, incorporating wireless communication among the electroacoustic components, allowing for freedom of movement during ANC operation. WNANC operates with a digital twin architecture, where adaptation of the control filter is managed by a cloud server, resolving the disadvantages of adaptive ANC in algorithm complexity and hardware cost. Nevertheless, the secondary path model remains crucial for WNANC. This paper introduces digital twin online secondary path modeling (DT-OSPM) algorithms for continuous secondary path estimation in WNANC. A key advantage of the DT-OSPM algorithms is their data-reuse capability, which decouples nested adaptive filters into sequential ones. Control filter adaptation occurs after the secondary path model is updated, resulting in improved convergence. Simulation and experiment results demonstrate the effectiveness and efficiency of the DT-OSPM algorithms in a feedforward WNANC system.

关键词： Noise Adaptive filters Digital twins Filtering algorithms Adaptation models Wireless communication Wireless sensor networks Feedforward systems Convergence Vectors

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：