检索结果-内蒙古大学图书馆

Multi-Scale Automated Self-Supervised Learning for Recommendation

学校读者我要写书评

暂无评论

Multi-Scale Automated Self-Supervised Learning for Recommend...

intelligent Management and Software Engineering (IMSE), International Conference on

作者： Yuhang Yan Liqin Tian Yujie Gao School of Computer Science Qinghai Normal University Xining China The State Key Laboratory of Tibetan Intelligent Information Processing and Application Xining China School of Computer North China Institute of Science and Technology Yanjiao China

Automated self-supervised learning frameworks have shown significant results in the recommendation domain. However, existing methods such as AutoCF use graph convolutional networks and graph attention mechanisms that are limited to a single expression granularity and cannot fully capture the multi-granularity of user interests and product attributes. To address this problem, we propose MS-AutoCF, a multi-scale automated self-supervised learning recommender system that obtains user and product representations at different granularities through multi-layer graph convolution to match the global and local details of their interest preferences. Meanwhile, we design a query key value-based multi-head graph attention module that can adaptively aggregate and enhance the representations with different granularities. Experimental results show that the MS- AutoCF system achieves significant improvements over existing single-scale methods on several public datasets.

关键词：

Dist-PU: Positive-Unlabeled Learning from a Label Distribution Perspective

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhao, Yunrui Xu, Qianqian Jiang, Yangbangyan Wen, Peisong Huang, Qingming School of Computer Science and Technology University of Chinese Academy of Sciences China Key Laboratory of Intelligent Information Processing Institute of Computing Technology CAS China State Key Laboratory of Information Security Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Laboratory of Big Data Mining and Knowledge Management University of Chinese Academy of Sciences China

Positive-Unlabeled (PU) learning tries to learn binary classifiers from a few labeled positive examples with many unlabeled ones. Compared with ordinary semi-supervised learning, this task is much more challenging due to the absence of any known negative labels. While existing cost-sensitive-based methods have achieved state-of-the-art performances, they explicitly minimize the risk of classifying unlabeled data as negative samples, which might result in a negative-prediction preference of the classifier. To alleviate this issue, we resort to a label distribution perspective for PU learning in this paper. Noticing that the label distribution of unlabeled data is fixed when the class prior is known, it can be naturally used as learning supervision for the model. Motivated by this, we propose to pursue the label distribution consistency between predicted and ground-truth label distributions, which is formulated by aligning their expectations. Moreover, we further adopt the entropy minimization and Mixup regularization to avoid the trivial solution of the label distribution consistency on unlabeled data and mitigate the consequent confirmation bias. Experiments on three benchmark datasets validate the effectiveness of the proposed method. Code available at: https://***/Ray-rui/Dist-PU-Positive-UnlabeledLearning-from-a-Label-Distribution-Perspective. Copyright © 2022, The Authors. All rights reserved.

关键词： Entropy

A CGRU multi-step wind speed forecasting model based on multi-label specific XGBoost feature selection and secondary decomposition

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Jiang, Zheyong Che, Jinxing School of Science Nanchang Institute of Technology Jiangxi Nanchang330090 China Jiangxi Province Key Laboratory of Water Information Cooperative Sensing and Intelligent Processing Nanchang Institute of Technology China

Accurate wind speed forecasting is beneficial to reduce the risk of power system caused by wind power uncertainty, which is of great significance for wind power grid connection. However, the randomness and intermittence of wind speed bring great challenges to the accurate wind speed forecasting. In this study, a CGRU multi-step wind speed forecasting model based on secondary decomposition (SD) and multi-label specific XGBoost feature selection method is proposed, which achieves good forecasting performance. By using the secondary decomposition and sample entropy analysis, the original wind speed series is decomposed into multiple sub-series, and the frequency division characteristics of wind speed fluctuations are further extracted. Previous studies rarely focused on feature selection for multi-step wind speed forecasting, the feature selection method suitable for multi-step wind speed forecasting is proposed in this study. By using this method, the optimal input features of each sub-series are selected, which simplifies the data structure and improves the modeling efficiency. An efficient CGRU forecasting model is developed in this study, which can capture the depth characteristics and obtain the time-dependent relationship of wind speed. Experiments conducted in four seasons show that the model can adapt to long-term dependence and extract effective information from original data. © 2022, The Authors. All rights reserved.

关键词： Feature extraction

FACOS: Enabling Privacy Protection Through Fine-Grained Access Control with On-chain and Off-chain System

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Chao Hou, Cankun Jiang, Tianyu Ning, Jianting Qiao, Hui Wu, Yusen Fujian Provincial Key Laboratory of Data Intensive Computing the Key Laboratory of Intelligent Computing and Information Processing Fujian China The School of Computer Science and Technology Xidian University Shaanxi Xi’an710071 China The Department of Computer Science University of Miami MiamiFL33146 United States College of Computer and Cyber Security Fujian Normal University Fuzhou350007 China The Faculty of Data Science City University of Macau China

Data-driven landscape across finance, government, and healthcare, the continuous generation of information demands robust solutions for secure storage, efficient dissemination, and fine-grained access control. Blockchain technology emerges as a significant tool, offering decentralized storage while upholding the tenets of data security and accessibility. However, on-chain and off-chain strategies are still confronted with issues such as untrusted off-chain data storage, absence of data ownership, limited access control policy for clients, and a deficiency in data privacy and auditability. To solve these challenges, we propose a permissioned blockchain-based privacy-preserving fine-grained access control on-chain and off-chain system, namely FACOS. We applied three fine-grained access control solutions and comprehensively analyzed them in different aspects, which provides an intuitive perspective for system designers and clients to choose the appropriate access control method for their systems. Compared to similar work that only stores encrypted data in centralized or non-fault-tolerant IPFS systems, we enhanced off-chain data storage security and robustness by utilizing a highly efficient and secure asynchronous Byzantine fault tolerance (BFT) protocol in the off-chain environment. As each of the clients needs to be verified and authorized before accessing the data, we involved the Trusted Execution Environment (TEE)-based solution to verify the credentials of clients. Additionally, our evaluation results demonstrated that our system1 offers better scalability and practicality than other state-of-the-art designs. We deployed our system on Alibaba Cloud and Tencent Cloud and conducted multiple evaluations. The results indicate that it takes about 2.79 seconds for a client to execute the protocol for uploading and about 0.96 seconds for downloading. Compared to other decentralized systems, our system exhibits efficient latency for both download and upload operations.

关键词： Access control

Artificial Intelligence-Generated Terahertz Multi-Resonant Metasurfaces via Improved Transformer and CGAN Neural Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Huang, Yangpeng Feng, Naixing Cai, Yijun Smart Sensing Integrated Circuit Engineering Research Center of Universities in Fujian Province Xiamen University of Technology Xiamen361000 China The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei230601 China Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui University Hefei230601 China

It is well known that the inverse design of terahertz (THz) multi-resonant graphene metasurfaces by using traditional deep neural networks (DNNs) has limited generalization ability. In this paper, we propose improved Transformer and conditional generative adversarial neural networks (CGAN) for the inverse design of graphene metasurfaces based upon THz multi-resonant absorption spectra. The improved Transformer can obtain higher accuracy and generalization performance in the StoV (Spectrum to Vector) design compared to traditional multilayer perceptron (MLP) neural networks, while the StoI (Spectrum to Image) design achieved through CGAN can provide more comprehensive information and higher accuracy than the StoV design obtained by MLP. Moreover, the improved CGAN can achieve the inverse design of graphene metasurface images directly from the desired multi-resonant absorption spectra. It is turned out that this work can finish facilitating the design process of artificial intelligence-generated metasurfaces (AIGM), and even provide a useful guide for developing complex THz metasurfaces based on 2D materials using generative neural networks. Copyright © 2023, The Authors. All rights reserved.

关键词： Graphene

Depth-Aware Multi-Modal Fusion for Generalized Zero-Shot Learning 22

学校读者我要写书评

暂无评论

Depth-Aware Multi-Modal Fusion for Generalized Zero-Shot Lea...

22nd IEEE International Conference on Industrial Informatics, INDIN 2024

作者： Cao, Weipeng Yao, Xuyang Xu, Zhiwu Pan, Yinghui Sun, Yixuan Li, Dachuan Qiu, Bohua Wei, Muheng Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Stony Brook University New York United States Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Shenzhen China Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China ZhenDui Industry Artificial Intelligence Co. Ltd Shenzhen China Department of Automation Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai China

ISBN: (纸本)9798331527471

Realizing Generalized Zero-Shot Learning (GZSL) based on large models is emerging as a prevailing trend. However, most existing methods merely regard large models as black boxes, solely leveraging the features output by the final layer while disregarding potential performance enhancements from other layers. Indeed, numerous researchers have visually depicted variations in the features learned across different layers of neural networks. Motivated by this observation, we propose a Vision Transformer (ViT)-based GZSL method named Depth-Aware Multi-Modal ViT (DAM2ViT), which exploits multi-level features of ViT. DAM2ViT incorporates a multi-modal interaction block to align semantic information of categories across multiple layers, thereby augmenting the model’s capacity to learn associations between visual and semantic spaces. Extensive experiments conducted on three benchmark datasets (i.e., CUB, SUN, AWA2) have showcased that DAM2ViT achieves competitive results compared to state-of-the-art methods. © 2024 IEEE.

关键词： Zero-shot learning

Electric Field Manipulation of Rydberg States for Very Low Frequency Fields Detection

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Chen, Minze Feng, Haonan Gao, Ge Zhu, Zhiao Li, Zhongxiang Wu, Zhonghuai Xiao, Wei Dai, Weidong Peng, Peng Zheng, Dezhi State Key Laboratory of Environment Characteristics and Effects for Near-Space Beijing Institute of Technology Beijing100081 China National Key Laboratory of Science and Technology on Space-Born Intelligent Information Processing Beijing100081 China State Key Laboratory of Explosion Science and Safety Protection Beijing Institute of Technology Beijing100081 China Jiaxing314019 China Zhuhai519088 China Advanced Research Institute of Multidisciplinary Science Beijing Institute of Technology Beijing100081 China Taizhou Meteorological Bureau Taizhou318000 China Chongqing Caeri Robot Technology Co. Ltd. Chongqing400722 China

The very low frequency(VLF) band is widely used in submarine communication and geophysical exploration for its strong penetration and long-distance propagation. This paper theoretically and experimentally investigates Rydberg EIT in 133Cs vapor under VLF and DC fields. A model is established to describe the EIT spectral response under dual-field conditions, with theoretical predictions showing agreement with experimental results. We propose a novel calibration-free method to measure VLF electric fields, bypassing traditional Stark shift measurements. This method detects additional splitting intervals of Stark sublevels, separated from the degenerate energy level under a DC field. This phenomenon arises from the averaging effect of sublevel sinusoidal oscillations in the spectrum induced by the VLF field. The splitting interval is proportionally dependent on the VLF field amplitude. The VLF electric field sensor is enhanced by increasing the strength of the DC field, extending the traceable measurement limit for weak VLF electric fields by more than an order of magnitude. This work highlights the potential for precise VLF electric field measurements, significantly advancing the calibration-free detection capabilities of Rydberg atom sensors for low-frequency applications. Copyright © 2025, The Authors. All rights reserved.

关键词： Rydberg states

Steel Surface Defect Detection Based on Improved YOLOv8

学校读者我要写书评

暂无评论

Steel Surface Defect Detection Based on Improved YOLOv8

International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)

作者： Jichen Zhang Zhi Zhang College of Computer Science and Technology Wuhan University of Science and Technology Wuhan Hubei China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan Hubei China Big Data Science and Engineering Research Institute Wuhan University of Science and Technology Wuhan Hubei China

ISBN: (数字)9798350356328

ISBN: (纸本)9798350356335

There are some defects on the surface of steel that are difficult to detect, so we propose an improved algorithm based on YOLOv8 for detecting defects that are difficult to be detected. This improvement includes the introduction of an SPD module to minimize the problem of information loss due to downsampling; the addition of a small object detection head combined with EMA attention to further enhance the detection capabilities for small targets; the use of BiFPN to replace the original PANet, applying different weights to different features to strengthen feature fusion. Experiments were conducted on the NEU-DET dataset, and the algorithm improved mAP to 80.2%, an improvement of 3.9% over the baseline model.

关键词： Convolution Computational modeling Object detection Feature extraction Knowledge discovery Steel Data mining Background noise Defect detection Fuzzy systems

Service-Driven Dynamic QoS On-Demand Routing Algorithm

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： She, Hao Yan, Lixing Mao, Chuanfeng Bu, Qihui Guo, Yongan College of Telecommunications & Information Engineering Nanjing University of Posts and Telecommunications Nanjing210003 China Jiangsu Key Laboratory of Intelligent Information Processing and Communication Technology Nanjing University of Posts and Telecommunications Nanjing210003 China Edge Intelligence Research Institute Nanjing University of Posts and Telecommunications Nanjing210003 China

With the proliferation of Internet of Things (IoT) devices, the scale of networks is growing exponentially. However, dynamically meeting the diverse quality of service (QoS) routing requirements for users and services in large-scale networks remains a critical challenge. To address this issue, this paper proposes a Service-Driven Dynamic QoS On-Demand model and establishes a corresponding QoS optimization objective function. The SHA-256 hash algorithm is utilized to simplify the large-scale network model, effectively reducing the number of Segment Routing (SR) nodes. The proposed Service-Driven Dynamic QoS On-Demand Routing Algorithm (SDDRL) identifies the optimal path, which is then uniformly disseminated by the SDN controller, thereby addressing existing challenges in SDN-IoT networks. Compared to OSPF-based and DDQN-based algorithms, the SDDRL algorithm reduces the average delay by 53.85\% and 31.63\%, respectively. The proposed algorithm improves the average network congestion degree and route calculation time compared to other existing algorithms, and it demonstrates superior performance in handling complex tasks. © 2024, The Authors. All rights reserved.

关键词： Quality of service