检索结果-内蒙古大学图书馆

Knowledge Graph-Enhanced Recommendation with Box Embeddings 23

23rd Chinese National Conference on Computational Linguistics, CCL 2024

作者： Liang, Qiuyu Wang, Weihua Lv, Lei Bao, Feilong College of Computer Science Inner Mongolia University Hohhot China National and Local Joint Engineering Research Center of Intelligent Information Processing Technology for Mongolian Hohhot China Inner Mongolia Key Laboratory of Mongolian Information Processing Technology Hohhot China

ISBN: (纸本)9780000000002

Knowledge graphs are used to alleviate the problems of data sparsity and cold starts in recommendation systems. However, most existing approaches ignore the hierarchical structure of the knowledge graph. In this paper, we propose a box embedding method for knowledge graph-enhanced recommendation system. Specifically, the box embedding represents not only the interaction between the user and the item, but also the head entity, the tail entity and the relation between them in the knowledge graph. Then the interaction between the item and the corresponding entity is calculated by the multi-task attention unit. Experimental results show that our method provides a large improvement over previous models in terms of Area Under Curve (AUC) and accuracy in publicly available recommendation datasets with three different domains. © 2024 China National Conference on Computational Linguistics.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Table-to-Text Generation with Structural Bias Attention 20th

Few-Shot Table-to-Text Generation with Structural Bias Atten...

引用

20th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2023

作者： Liu, Di Wang, Weihua Bao, Feilong Gaov, Guanglai College of Computer Science Inner Mongolia University Hohhot China National and Local Joint Engineering Research Center of Intelligent Information Processing Technology for Mongolian Hohhot China Inner Mongolia Key Laboratory of Mongolian Information Processing Technology Hohhot China

ISBN: (纸本)9789819970216

Table-to-text generation task refers to converting tabular data into language text to facilitate easier understanding and analysis of the table. Recently, pre-trained models have made significant advancements in this kind of tasks. However, the inherent structural differences between tabular data and text, and the lack of domain-specific knowledge in few-shot datasets, make it challenging for pre-trained models to generate faithful text. To solve these problems, we proposed a framework that encodes tables by obtaining structural bias attention through pruning full self-attention, distinguishing the importance of cells from a structural perspective. We use the pre-trained model with the structural bias framework to the generation component of Prototype-to-Generation. To encourage prototype memory to adhere to the table content and generate more accurate and aligned sentences, we employ Reinforcement Learning. We conducted extensive experiments on three few-shot table datasets. Compared to previous advanced methods, our model achieved superior performance across multiple evaluation metrics. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Wide Stopband Substrate Integrated Waveguide Filter Using Bisection and Trisection Coupling in Multilayer

引用

Chinese Journal of Electronics 2024年第2期33卷 436-442页

作者： Peng CHU Jianguo FENG Peng ZHU Lei GUO Long ZHANG Leilei LIU Guoqing LUO Ke WU College of Electronic and Optical Engineering Nanjing University of Posts and Telecommunications National and Local Joint Engineering Laboratory of RF Integration and Micro-Assembly Technology School of Information and Communication Engineering Dalian University of Technology Guangdong Provincial Key Laboratory of Optoelectronic Information Processing Chips and Systems Sun Yat-sen University College of Electronics and Information Engineering Shenzhen University School of Electronics and Information Hangzhou Dianzi University Poly-Grames Research Center Polytechnique Montreal

This article presents a highly efficient method for substrate-integrated-waveguide（SIW） filters to achieve very wide stopbands. By employing the proposed trisection slots in addition to the bisection slots as the inter-coupling structures, all spurious modes below TE505 of a SIW filter working in the fundamental mode TE101（f0）can be eliminated without requiring additional structure or complex theoretical analysis, without affecting the design of the fundamental passbands, and without degrading the performance of the filters. For verification, two prototype filters are designed, fabricated, and measured with wide stopbands up to 4.15f0 and 4.83f0, respectively. The proposed technique could facilitate the development of high-performance wide-stopband SIW filters for microwave/wireless circuits and systems.

关键词： Bandpass filter Wide stopband Substrate integrated waveguide Spurious mode suppression Multilayer

来源：评论

学校读者我要写书评

暂无评论

Improving ECAPA-TDNN Performance with Coordinate Attention

引用

Journal of Shanghai Jiaotong University (Science) 2024年 1-7页

作者： Liu, Shuanghong Song, Zhida He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China Department of Electronic Engineering and Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

The current mainstream networks, such as squeeze and excitation residual neural network (SE-ResNet) and emphasized channel attention, propagation and aggregation based time delay neural network (ECAPA-TDNN), enhance the capability of speaker embedding extractors to extract more discriminative speaker embeddings by incorporating squeeze and excitation (SE) attention within the convolutional blocks. However, the SE attention focuses solely on encoding inter-channel information, overlooking the importance of spatial positional information and time-frequency information, which are crucial for the model’s performance. In this paper, we first experimentally compare the effectiveness of several mainstream attention mechanisms in the computer vision domain for the ECAPA-TDNN model. Next, we focus on the substantial improvements that coordinate attention (CA) brings to the ECAPA-TDNN model. The introduction of CA can help the model embed time-frequency information into the channel representation. Even without using AS-Norm, our proposed model achieves relative reductions of about 5.3% equal error rate (EER) and 5.5% minimum detection cost function (minDCF) on both the Voxceleb-O and Voxceleb-H test sets compared to the ECAPA-TDNN baseline model. In addition, the EER is relatively reduced by 9.46% on the CN-Celeb1 test set. This result strongly demonstrates that the CA module can effectively improve the generalization ability of the ECAPA-TDNN model. © Shanghai Jiao Tong University 2024.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

UAV-Assisted Dynamic Avatar Task Migration for Vehicular Metaverse Services: A Multi-Agent Deep Reinforcement Learning Approach

引用

IEEE/CAA Journal of Automatica Sinica 2024年第2期11卷 430-445页

作者： Jiawen Kang Junlong Chen Minrui Xu Zehui Xiong Yutao Jiao Luchao Han Dusit Niyato Yongju Tong Shengli Xie IEEE School of Automation Guangdong University of TechnologyGuangzhou 510006China 111 Center for Intelligent Batch Manufacturing based on IoT Technology Guangzhou 510006China Key Laboratory of Intelligent Information Processing and System Integration of IoT Ministry of EducationGuangzhou 510006China School of Computer Science and Engineering Nanyang Technological UniversitySingaporeSingapore Pillar of Information Systems Technology and Design Singapore University of Technology and DesignSingaporeSingapore College of Communication Engineering Army Engineering University of PLANanjing 210007China National Natural Science Foundation of China Beijing 100085China Key Laboratory of Intelligent Detection and IoT in Manufacturing Ministry of EducationGuangzhou 510006China Guangdong Key Laboratory of IoT Information Technology Guangzhou 510006China

Avatars, as promising digital representations and service assistants of users in Metaverses, can enable drivers and passengers to immerse themselves in 3D virtual services and spaces of UAV-assisted vehicular Metaverses. However, avatar tasks include a multitude of human-to-avatar and avatar-to-avatar interactive applications, e.g., augmented reality navigation,which consumes intensive computing resources. It is inefficient and impractical for vehicles to process avatar tasks locally. Fortunately, migrating avatar tasks to the nearest roadside units(RSU)or unmanned aerial vehicles(UAV) for execution is a promising solution to decrease computation overhead and reduce task processing latency, while the high mobility of vehicles brings challenges for vehicles to independently perform avatar migration decisions depending on current and future vehicle status. To address these challenges, in this paper, we propose a novel avatar task migration system based on multi-agent deep reinforcement learning(MADRL) to execute immersive vehicular avatar tasks dynamically. Specifically, we first formulate the problem of avatar task migration from vehicles to RSUs/UAVs as a partially observable Markov decision process that can be solved by MADRL algorithms. We then design the multi-agent proximal policy optimization(MAPPO) approach as the MADRL algorithm for the avatar task migration problem. To overcome slow convergence resulting from the curse of dimensionality and non-stationary issues caused by shared parameters in MAPPO, we further propose a transformer-based MAPPO approach via sequential decision-making models for the efficient representation of relationships among agents. Finally, to motivate terrestrial or non-terrestrial edge servers(e.g., RSUs or UAVs) to share computation resources and ensure traceability of the sharing records, we apply smart contracts and blockchain technologies to achieve secure sharing management. Numerical results demonstrate that the proposed approach

关键词： Avatar blockchain metaverses multi-agent deep reinforcement learning transformer UAVs

来源：评论

学校读者我要写书评

暂无评论

Topology Uncertainty Modeling For Imbalanced Node Classification on Graphs 48

Topology Uncertainty Modeling For Imbalanced Node Classifica...

引用

48th IEEE International Conference on Acoustics, Speech and Signal processing, ICASSP 2023

作者： Gao, Jiayi Li, Jiaxing Zhang, Ke Kong, Youyong Southeast University Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing School of Computer Science and Engineering Nanjing China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing China

ISBN: (纸本)9781728163277

Most existing graph neural networks work under a class-balanced assumption, while ignoring class-imbalanced scenarios that widely exist in real-world graphs. Although there are many methods in other fields that can alleviate this issue, they do not consider the special topology of the non-Euclidean graph. Hence, we propose Graph Topology Uncertainty (GraphTU), a novel probabilistic class-imbalanced solution specifically for graphs. Firstly, an invisible "uncertain gap"between under-represented minorities in training set and authentic minorities in unseen set is modeled by estimating statistical variances in topology. We extend the training distribution for minorities by sampling in this gap through a non-parametric way. Moreover, a gradient-guided mask is introduced to prevent biased statistics. Extensive experiments demonstrate the superior performance of GraphTU. © 2023 IEEE.

关键词： Class Imbalance Node Classification

来源：评论

学校读者我要写书评

暂无评论

Topgformer: Topological-Based Graph Transformer for Mapping Brain Structural Connectivity to Functional Connectivity 48

Topgformer: Topological-Based Graph Transformer for Mapping ...

引用

48th IEEE International Conference on Acoustics, Speech and Signal processing, ICASSP 2023

作者： Guo, Dalu Zhang, Ke Li, Jiaxing Kong, Youyong Southeast University Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing School of Computer Science and Engineering Nanjing China Southeast University Ministry of Education Key Laboratory of Computer Network and Information Integration Nanjing China

ISBN: (纸本)9781728163277

Exploring the mapping between structural connectivity (SC) and functional connectivity (FC) is of essential importance to understanding the working mechanism of the human brain. Traditional methods are difficult to represent the complex relationship of high-order interaction between SC and FC. Recent learning-based methods can not well capture the important long-range interactions and edge information of brain connectivity. To address the issue, we propose a novel Topological-based Graph Transformer (TopGFormer) to generate functional connectivity from the structure connectivity with sufficient consideration of topological properties of brain connectivity. We propose a Topological Multi-Head Attention block that simultaneously uses centrality encoding and adjacency matrix to capture node and edge importance respectively. The centrality encoding reflects the importance of brain regions while the adjacency matrix captures the linking relationship of different regions. Experiments on HCP resting and emotion task datasets demonstrate the performance of the proposed method. © 2023 IEEE.

关键词： functional connectivity structural connectivity Structure-function mapping transformer

来源：评论

学校读者我要写书评

暂无评论

Multi-stage Image Deraining based on Pre-trained Diffusion Model 6

Multi-stage Image Deraining based on Pre-trained Diffusion M...

引用

6th ACM International Conference on Multimedia in Asia, MMAsia 2024

作者： Zeng, Xiong Jiang, Min Huang, RongHua School of Computer Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology Hubei Wuhan China

ISBN: (纸本)9798400712739

Image deraining typically involves synthesizing low-quality degraded data for training using a predefined degraded model of a single weather condition. While in real world scenarios, varying rain intensities result in different sizes and densities of raindrops and rain streaks, increasing the complexity of image degradation. In this paper, we proposed a multi-stage deraining framework based on pre-trained diffusion model, it can efficiently perform the rain removal task under a variety of weather situations. We diffuse degraded images into a noisy state where various types of degradation are transformed into Gaussian noise. Then, during the denoising process, the low-frequency information of the image is replaced through iterative refinement, guiding the pre-trained diffusion model for image reconstruction. Our method effectively utilizes the generative priors in diffusion models and avoid the computational burden of retraining conditional diffusion models. Experimental results on four rainy degradation image datasets show its robustness to different types and severities of degradation (such as raindrops and rain streaks). Compared to recent deraining algorithms, our method achieves a maximum improvement of 0.96 dB (3.5%) in PSNR and 0.021 dB (2.7%) in SSIM for the restored images. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Rain

来源：评论

学校读者我要写书评

暂无评论

Time series classification based on Unet with 2D transformation 12

Time series classification based on Unet with 2D transformat...

引用

12th IEEE International Conference on Identification, information and Knowledge in the Internet of Things, IIKI 2024

作者： Du, Mingsen Wang, Yunxia Wei, Yanxuan Wu, Xiaona Ji, Cun Shandong Normal University School of Information Science and Engineering Jinan China Shandong Provincial Key Laboratory for Distributed Computer Software Novel Technology Jinan China

ISBN: (纸本)9798331510633

Time series classification has gained significant attention in the data mining field in recent years. Despite the proposal of numerous methods over the past decades, most of these approaches focus on feature extraction in one-dimensional space, which often results in lower classification accuracy. To address this issue, this paper proposes a Time Series Classification method based on U-Net with 2D transformation (TSC-Unet). This method first transforms univariate time series into 2D space and augments the data, effectively enhancing the diversity of the dataset. Subsequently, features are extracted from the 2D space using a fully convolutional network, fully exploiting multi-scale feature information. U-Netwith skip connections, effectively retains and transmits high-resolution features, significantly reducing information loss and improving segmentation performance. Additionally, the ability to learn multi-scale features enables the model to capture details at different levels, thereby enhancing its understanding of complex data. Experimental results demonstrate that TSC-Unet achieves high classification accuracy on the UCR datasets, validating the effectiveness and superiority of the proposed method. © 2024 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

MASS FLOW DETECTION technology FOR SEED AND FERTILISER PARTICLES AND ITS APPLICATION IN UAV-BASED SPREADING

引用

INMATEH - Agricultural Engineering 2025年第1期75卷 584-599页

作者： Wang, Yanshuang Song, Cancan Gao, Yongqi Lan, Yubin Wang, Guobin Ming, Rui College of Agricultural Engineering and Food Science Shandong University of Technology Zibo China Shandong University of Technology Sub-center of National Center for International Collaboration Research on Precision Agricultural Aviation Pesticide Spraying Technology Zibo China Research of Institute of Ecological Unmanned Farm Shandong University of Technology Zibo China School of Computer and Big Data Minjiang University Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Fuzhou China

The processes of sowing and fertiliser application represent a significant aspect of agricultural production. In order to achieve efficient and precise seeding and fertiliser application, mass flow detection of seed and fertiliser particles can facilitate real-time monitoring and precise decision-making for intelligent seeding and fertiliser application. However, the diversity of seed and fertiliser particle types and particle flow modes presents a challenge for existing detection methods, particularly in meeting the varying operational requirements, especially for particle mass flow detection in low-altitude and high-speed sowing operations of agricultural drones. In such cases, the overlap and flow rate will have a significant impact on the detection results due to the large displacement and the continuous high-throughput dense phase of the particle flow. This paper provides a summary of the existing seed and fertiliser particle mass flow detection techniques and their underlying working principles. It compares the direct detection based on mass with the indirect detection methods based on velocity and concentration, and analyses their respective advantages, disadvantages and applicability. It also explores the possibility of optimising the existing detection methods for the specific needs of agricultural UAVs and considers the potential introduction of cutting-edge science and technology in order to develop an efficient, accurate and convenient detection system to meet the growing market demand. © (2025), (INMA Bucharest). All Rights Reserved.

关键词： Precision agriculture

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：