检索结果-内蒙古大学图书馆

15th IEEE International Conference on ASIC, ASICON 2023

作者： Li, Yuxin Duan, Zhenxia Wang, Jie Zhou, Zijian Yang, Ming Wu, Enqi Du, Lin University of Shanghai for Science and Technology School of Mechanical Engineering Shanghai China University of Shanghai for Science and Technology School of Optical-Electrical and Computer Engineering Shanghai200433 China

ISBN: (纸本)9798350312980

The advancement of semiconductor fabrication methods has sped up the development of bionic structures and biosensing technologies. The device structure of an imitation lotus leaf and the digital biomolecular diagnostic technology are developed in this work using the wet etching procedure. The mold formed of high borosilicate glass BF33 has the benefit of low roughness and superior etching controllability over a calcium sodium glass, and the roundness parameter is raised by an average of more than 4% up to an etching depth of more than 100 μm. The produced droplet volume by device preparation can have a relative standard deviation as low as 4.5%, and the one-time filling rate is greater than 90%. The novel tool is used in biological Loop-Mediated Isothermal Amplification (LAMP) assessment to confirm that digital detection and biocompatibility are feasible. © 2023 IEEE.

关键词： BF33 glass droplet array LAMP wet deep etching

来源：评论

学校读者我要写书评

暂无评论

Explainable Natural Language Inference via Identifying Important Rationales

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第3期4卷 438-449页

作者： Yang, Zongbao Dong, Shoubin Hu, Jinlong South China University of Technology School of Computer Science and Engineering Guangzhou510006 China Zhongshan Institute of Modern Industrial Technology of Scut Zhongshan528437 China

Natural language inference (NLI) is an important task in the field of natural language processing (NLP), which requires certain common sense and logical reasoning abilities. Existing pretrained models applied to NLI have difficulty achieving good performance, and inference models based on deep learning also have the problem of poor interpretability. In this article, we propose an explainable NLI model by identifying important rationales, which can be obtained by the specially designed rationale extractor and selector. The pretrained models can be fine-tuned on the rationale-enhanced dataset. In addition, the extracted rationales can be used to generate natural language explanation (NLE) with a higher quality. Not only can NLE serve as an explanation for the prediction results, but it can also be combined with the original dataset to improve the performance of pretrained models. Experimental results on the Stanford natural language inference corpus (SNLI) show that the proposed model could obtain state-of-the-art results of 94.19% and 94.08%, respectively, on the SNLI development set, as well as on the test set and obtain good performance when transferred to the out-of-domain dataset (multiNLI). Furthermore, the proposed model has a certain improvement in computational efficiency due to the focus on important rationales. © 2020 IEEE.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Improving Dynamic Behavior of Vehicular Ad Hoc Networks by Integrating Game Theory Technique 5th

Improving Dynamic Behavior of Vehicular Ad Hoc Networks by I...

引用

5th International Conference on Applied Technologies, ICAT 2023

作者： Tiwari, Shobhit Mani Baghel, Anurag Singh Department of Computer Science and Engineering Faculty of Engineering and Technology University of Lucknow Uttar Pradesh Lucknow India Department of Computer Science and Engineering School of ICT Gautam Buddha University Uttar Pradesh Greater Noida India

ISBN: (纸本)9783031589553

In this paper, we give the focus on the continuous advancement in the domain of Vehicular ad hoc networks (VANET’s) and that is developed as a tool for developing the base for platform intelligent mode in the communication (inter vehicle communications) and needful for enhanced improvement under the performance effect and traffic safety. Here, we have deployed the game theory method in the VANET domain and thereafter categorizes between the hop vehicle and the source vehicle. The New approach to the VANET analysis we have identified and used the social parameters and hence made it look more novel approach. "Researching vehicle ad hoc networks poses challenges due to their dynamic nature, the movement of large vehicles, unlimited energy resources, and the evolution of wireless networks. Game theory is commonly employed in wireless networks to explore the interplay between competition and cooperation. In this study, we present a system for ad hoc networks in cars, utilizing game theory to automate car groups and board elections. This approach eliminates the necessity for regular bulk updates. Furthermore, the social behavior of each car is leveraged to create clusters in the car environment. The K-means algorithm in machine learning is applied to develop social cars. The proposed system underwent testing for various characteristics, including CH lifetime, average group lifetime, average number of joins, throughput, and packet loss rate. The results demonstrate that VANET performs exceptionally well, achieving an overall performance ranging from 0.95 to 0.989". © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Heterogeneous networks

来源：评论

学校读者我要写书评

暂无评论

Locating What You Need: Towards Adapting Diffusion Models to OOD Concepts In-the-Wild 38

Locating What You Need: Towards Adapting Diffusion Models to...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Yang, Jianan Gao, Chenchao Xiao, Zhiqing Zhao, Junbo Wu, Sai Chen, Gang Wang, Haobo College of Computer Science and Technology Zhejiang University China Institute of Blockchain and Data Security China School of Software Technology Zhejiang University China International School of Information Science and Engineering Dalian University of Technology China

The recent large-scale text-to-image generative models have attained unprecedented performance, while people established adaptor modules like LoRA and DreamBooth to extend this performance to even more unseen concept tokens. However, we empirically find that this workflow often fails to accurately depict the out-of-distribution concepts. This failure is highly related to the low quality of training data. To resolve this, we present a framework called Controllable Adaptor Towards Out-of-Distribution Concepts (CATOD). Our framework follows the active learning paradigm which includes high-quality data accumulation and adaptor training, enabling a finer-grained enhancement of generative results. The aesthetics score and concept-matching score are two major factors that impact the quality of synthetic results. One key component of CATOD is the weighted scoring system that automatically balances between these two scores and we also offer comprehensive theoretical analysis for this point. Then, it determines how to select data and schedule the adaptor training based on this scoring system. The extensive results show that CATOD significantly outperforms the prior approaches with an 11.10 boost on the CLIP score and a 33.08% decrease on the CMMD metric. © 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

An adaptive data hiding for AMBTC compressed images with recoverability using bound shifting technique

引用

Multimedia Tools and Applications 2023年第10期82卷 15593-15612页

作者： Zhou, Xiaoyu Yang, Guangsong Hong, Wien Chen, Kia Sheng Chen, Tung-Shou School of Ocean Information Engineering Jimei University Xiamen361021 China Department of Computer Science and Information Engineering National Taichung University of Science and Technology Taichung City404 Taiwan School of Electrical and Computer Engineering Nanfang College of Guangzhou Guangzhou510970 China

This paper proposes an efficient data hiding method for absolute moment block truncation coding (AMBTC) images with the recoverability of compressed code. The existing methods sacrifice some embedding capacity and image quality for recovering the AMBTC compressed code. To remedy these problems, this paper divides the quantized differences into four intervals to embed data bits with different lengths. Therefore, the embedding capacity is improved. Moreover, our method uses the bound shifting technique (BST) to provide a significant improvement in image quality. The experimental results show that the payload of our method increases by 19.14% and the image quality is also enhanced by 8.02% compared with prior works for the test Lena image. © 2022, The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature.

关键词： Image quality

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Analysis of Machine Learning and Deep Learning Based Product Recommendation System 5th

A Comprehensive Analysis of Machine Learning and Deep Learni...

引用

5th EAI International Conference on Cognitive Computing and Cyber Physical Systems, IC4S 2024

作者： Chakravarthi, GKalyan Kumar, Raghvendra Addagarla, Ssvr Kumar Department of Computer Science and Engineering School of Engineering and Technology GIET University Odisha Gunupur India Nadimpalli Satyanarayana Raju Institute of Technology Visakhapatnam India

ISBN: (纸本)9783031770746

Recommender Systems (RS) have been widely applied in various real-time applications to support identifying valuable information. The RS tries to give actual suggestions to every user based on their behavior as well as interests. Recommendations generated by these systems often contend with the unique personal interests of individual users, thereby playing a pivotal role in their decision-making processes. Recommender Systems (RS) act as efficient tools for filtering vast amounts of online data, shaping the behaviors of smartphone users, personalization trends, and the evolution of internet access. RS are generally classified into three primary types: Hybrid Filtering, Content-Based Filtering (CBF), and Collaborative Filtering (CF). These systems find wide application across various fields including online education systems, e-commerce, marketing, tourism, food service, movies, business, and beyond. Although the recent RS are well-known in providing valuable recommendations, they suffer from a number of limitations as well as challenges such as scalability, sparsity, and cold-start and so on. Due to the various approach’s existence, the selection of these approaches becomes challenging during the development of application-based RS. Moreover, every approach comes with its individual feature sets, advantages as well as limitations, which must be addressed. This survey reviews the research inclinations which integrates the progressive technical characteristics of the RS. The aim of this survey is to ensure a systematic review on recent contributions in RS domain, and concentrated on various applications such as education, food, products and so on. This survey provides a comprehensive review of these types of RS, recent literature, and applications of visual RS. © ICST Institute for computer sciences, Social Informatics and Telecommunications engineering 2025.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Maximizing the Accuracy of Time Series Analysis for the Early Detection of Rhabdomyosarcomas

Maximizing the Accuracy of Time Series Analysis for the Earl...

引用

2024 International Conference on Optimization Computing and Wireless Communication, ICOCWC 2024

作者： Shree, Ritu Prabhu, A. Gautam, Anupam Kumar Vivekananda Global University Department of Computer Science & Application Jaipur India Department of Computer Science and Information Technology Bangalore India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India

ISBN: (纸本)9798350383348

Rhabdomyosarcomas are a rare malignant tender tissue tumor that generally gives in younger children and teenagers. Early prognosis and remedy are essential for successful outcomes. Time collection evaluation is a valuable tool for recognizing styles and trends in medical facts, mainly for rare situations, which include Rhabdomyosarcomas. It has consequently been increasingly employed to detect early signs and symptoms of the ailment. On this look, we are conscious of investigating and optimizing techniques for time collection analysis. It is an excellent way to enhance its application and accuracy in identifying early symptoms and signs of rhabdomyosarcoma. We examine present strategies and suggest improvement techniques, along with function extraction and system mastering techniques. We further inspect the effectiveness of our strategies by conducting experiments on a dataset installed from scientific facts and literature of rhabdomyosarcoma instances. Those experiments show promising effects, indicating that our proposed strategies can considerably increase the accuracy and sensitivity of time series evaluation for the early detection of rhabdomyosarcoma and cause higher prognoses for affected sufferers. The focal point of this study is to maximize the accuracy of time series analysis for the early detection of rhabdomyosarcoma. Time collection analysis includes: • The gathering of temporal information from multiple sources. • The assessment of these records. • The interpretation of correlations between the facts points. This study aims to utilize these techniques to discover diffused adjustments in affected person information so that you can perceive the onset of the disorder in advance than would be possible with traditional techniques. The examination will expand algorithms to systematically and accurately procedure the temporal statistics and discover adjustments indicative of Rhabdomyosarcomas. In addition, the look will rent gadget learning to boost the dete

关键词： malignant recognizing Rhabdomyosarcomas strategies Time

来源：评论

学校读者我要写书评

暂无评论

Integrating Star Graph and Dynamic Intent Awareness for Session-based Recommendation 6

Integrating Star Graph and Dynamic Intent Awareness for Sess...

引用

6th International Conference on computer Network, Electronic and Automation, ICCNEA 2023

作者： Mao, Mingyi Xu, Jiayi Beijing Technology and Business University School of Computer Science and Engineering Beijing China

ISBN: (纸本)9798350305388

A session-based recommendation model combining star graph and dynamic perception is proposed to solve the problem of remote information, ignoring the dynamic change of users' interests and the inaccurate expression of short-Term preferences. Construct star graph and obtain remote node information by using Graph neural network. Then, the dual gating mechanism is used to integrate the information of item, time and user, and establish the dynamic interest model of user. Improved adaptive fusion mechanism, combining remote node information with users' dynamic interests to get global preferences. Improve short-Term graphs and enhance recent interactive representations with global information for more accurate short-Term interest. Compared with the optimal benchmark model, HR@20 reaches 54.29% at the highest, an increase of more than 2.1 percentage points, and MRR@20 reaches 18.84% at the highest, an increase of more than 0.9 percentage points, which verifies that this method can better match the real intention of users. © 2023 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Unsigned Road Incidents Detection Using Improved RESNET From Driver-View Images

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第5期6卷 1203-1216页

作者： Li, Changping Wang, Bingshu Zheng, Jiangbin Zhang, Yongjun Chen, C.L. Philip Northwestern Polytechnical University School of Software Xi’an710129 China Shenzhen University Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen Key Laboratory of Media Security Shenzhen518060 China Guizhou University State Key Laboratory of Public Big Data College of Computer Science and Technology Guiyang550025 China South China University of Technology School of Computer Science and Engineering Guangzhou510641 China

Frequent road incidents cause significant physical harm and economic losses globally. The key to ensuring road safety lies in accurately perceiving surrounding road incidents. However, the highly dynamic nature of traffic introduces significant challenges, particularly in detecting sudden and temporary incidents. In this article, we propose a novel detection framework, multihead attention ResNet with dynamic bottleneck (DB_RESNET_MHA), to identify physical unsigned road incidents. Our approach introduces three key innovations. First, we develop a tailored data augmentation strategy to create images that closely mimic the complex variations found in real-world road environments. Second, we enhance model expressiveness by employing attention mechanisms to nonlinearly integrate convolutional kernels within the residual network. Furthermore, we refine the prediction head by applying spatially distinct attention weights, enabling the model to capture intricate correlations between different features more effectively. To demonstrate the effectiveness of our method, we create a dataset for unsigned road incidents (UnsignRI), comprising a total of 16 323 images that capture 12 distinct types of incidents. It stands out as the most comprehensive dataset in the field, encompassing a wide range of geographical features and incident categories. Experimental results show that DB_RESNET_MHA achieves an average accuracy of 96.2% and a f1-score of 0.955 across various categories of unsigned incidents, surpassing other models. © 2020 IEEE.

关键词： Highway accidents

来源：评论

学校读者我要写书评

暂无评论

Channel assignment and power allocation for throughput improvement with PPO in B5G heterogeneous edge networks

引用

Digital Communications and Networks 2024年第1期10卷 109-116页

作者： Xiaoming He Yingchi Mao Yinqiu Liu Ping Ping Yan Hong Han Hu The College of Computer and Information Hohai UniversityNanjingChina The Key Laboratory of Water Big Data Technology of Ministry of Water Resources Hohai UniversityNanjing China The School of Computer Science and Engineering Nanyang Technological UniversitySingapore The College of Textile and Clothing Engineering Soochow UniversitySuzhouChina The College of Internet of Things Nanjing University of Posts and TelecomurunicationsNanjingChina

In Beyond the Fifth Generation(B5G)heterogeneous edge networks,numerous users are multiplexed on a channel or served on the same frequency resource block,in which case the transmitter applies coding and the receiver uses interference ***,uncoordinated radio resource allocation can reduce system throughput and lead to user inequity,for this reason,in this paper,channel allocation and power allocation problems are formulated to maximize the system sum rate and minimum user achievable *** the construction model is non-convex and the response variables are high-dimensional,a distributed Deep Reinforcement Learning(DRL)framework called distributed Proximal Policy Optimization(PPO)is proposed to allocate or assign ***,several simulated agents are trained in a heterogeneous environment to find robust behaviors that perform well in channel assignment and power ***,agents in the collection stage slow down,which hinders the learning of other ***,a preemption strategy is further proposed in this paper to optimize the distributed PPO,form DP-PPO and successfully mitigate the straggler *** experimental results show that our mechanism named DP-PPO improves the performance over other DRL methods.

关键词： B5G Heterogeneous edge networks PPO Channel assignment Power allocation Throughput

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：