检索结果-内蒙古大学图书馆

GSRDR-GAN: Global search result diversification ranking approach based on multi-head self-attention and GAN

Neurocomputing 2025年 648卷

作者： Weidong Liu Jinzhong Li Shengbo Chen College of Software Henan University Kaifeng 475000 China School of Electronic and Information Engineering Jinggangshan University Ji’an 343009 China Key Laboratory of Embedded System and Service Computing (Tongji University) Ministry of Education Shanghai 201804 China Jiangxi Provincial Key Laboratory of Electronic Data Control and Forensics Jinggangshan University Ji’an 343009 China School of Computer and Information Engineering Henan University Kaifeng 475000 China

Search result diversification ranking aims to generate rankings that comprehensively cover multiple subtopics, but existing methods often struggle to balance ranking diversity with relevance and face challenges in modeling document interactions and dealing with limited high-quality training data. While GAN have proven highly successful in fields like computer vision, their application to search result diversification has been limited due to the discrete nature of ranking items and the complex interactions among documents. To address these challenges, we propose GSRDR-GAN, a novel approach that integrates multi-head self-attention with GAN. Our method consists of four key components designed to address the limitations of traditional approaches: the Selected Document State Retriever, the Subtopic Encoder with Multi-head Self-Attention, the Subtopic Decoder with Multi-head Self-Attention, and the Relevance Predictor. First, a self-attention-based feature extraction module is employed to enhance document representations, enabling the model to capture both global and local context effectively. Second, a GAN framework is introduced to improve generalization by generating diverse rankings, mitigating limited high-quality training data. Third, a carefully designed reward function optimizes the trade-off between ranking diversity and relevance, allowing the model to adaptively prioritize these competing objectives during training. Notably, the method improves the generator’s stability and the diversity of search results by reducing training variance, even without pre-trained models. Extensive experiments on the TREC Web Track dataset demonstrate that the proposed GSRDR-GAN method significantly enhances result diversity, achieving relative improvements of 1.7% in α -nDCG, 3.0% in ERR-IA, 3.3% in NRBP, and 0.9% in S-rec over strong baseline methods. Ablation studies and comparative analyses of different reward computation methods further validate the effectiveness of the propo

关键词：

来源：评论

学校读者我要写书评

暂无评论

Variable petri nets for mobility

arXiv

引用

arXiv 2021年

作者： Ding, Zhijun Yang, Ru Cui, Puwen Zhou, MengChu Jiang, Changjun Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai201804 China Department of Computer Science and Technology Tongji University Shanghai201804 China Institute of Systems Engineering Macau University of Science and Technology Macau999078 China Department of Electrical and Computer Engineering New Jersey Institute of Technology NewarkNJ07102 United States

Mobile computing systems, service-based systems and some other systems with mobile interacting components have recently received much attention. However, because of their characteristics such as mobility and disconnection, it is difficult to model and analyze them by using a structure-fixed model. This work proposes a new Petri net model called Variable Petri Net (VPN) for modeling and analyzing these systems. The definition, firing rule, and related analysis technology of VPN are introduced in detail. In a VPN, the possible interaction interfaces are abstracted as a new kind of places called virtual places, and the occurrences of (dis)connections are described by new functions, which makes it appropriate to describe the component collaboration in systems and realize the scalability and pluggability of systems. Moreover, to overcome the shortcoming that markings cannot reflect link capability of a system, VPNs add a constraint function along with a marking to represent a complete system configuration. Several examples are used to demonstrate the newly proposed model and method. Copyright © 2021, The Authors. All rights reserved.

关键词： embedded systems

来源：评论

学校读者我要写书评

暂无评论

Base station network traffic prediction approach based on LMA-DeepAR

arXiv

引用

arXiv 2021年

作者： Zhang, Jiachen Zuo, Xingquan Xu, Mingying Han, Jing Zhang, Baisheng Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education Beijing Uinversity of Posts and Telecommunications Beijing China Key Laboratory of Intelligent Telecommunication Software and Multimedia Beijing Uinversity of Posts and Telecommunications Beijing China Network Interface Virtualization Shanghai System Design Deptment Zhongxing Telecommunication Equipment Corporation Shanghai China

Accurate network traffic prediction of base station cell is very vital for the expansion and reduction of wireless devices in base station cell. The burst and uncertainty of base station cell network traffic makes the network traffic nonlinear and non-stationary, which brings challenges to the long-term prediction of network traffic. In this paper, the traffic model LMA-DeepAR for base station network is established based on DeepAR. Acordding to the distribution characteristics of network traffic, this paper proposes an artificial feature sequence calculation method based on local moving average (LMA). The feature sequence is input into DeepAR as covariant, which makes the statistical characteristics of network traffic near a period of time in the past be considered when updating parameters, and the interference of non-stationary network traffic on model training will be reduced. Experimental results show that the proposed prediction approach (LMA-DeepAR) outperforms other methods in the overall long-term prediction performance and stability of multi cell network traffic. Copyright © 2021, The Authors. All rights reserved.

关键词： Base stations

来源：评论

学校读者我要写书评

暂无评论

Learning How to Avoiding Obstacles for End-to-End Driving with Conditional Imitation Learning 19

Learning How to Avoiding Obstacles for End-to-End Driving wi...

引用

Proceedings of the 2019 2nd International Conference on Signal Processing and Machine Learning

作者： Enwei Zhang Hongtu Zhou Yongchao Ding Junqiao Zhao Chen Ye The Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University School of Electronics and Information Engineering Shanghai P. R. China Tongji University School of Electronics and Information Engineering Shanghai P. R. China Tongji University School of Electronics and Information Engineering P. R. China The Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University School of Electronics and Information Engineering P. R. China

ISBN: (纸本)9781450372213

Obstacle avoiding is one of the most complex tasks for autonomous driving systems, which was also ignored by many cutting-edge end-to-end learning-based methods. The difficulties stem from the integrated process of detection and interpretation of environment and obstacles and generation of proper behaviors. We make the use of CARLA, a simulator for autonomous driving research, and collect massive human drivers' reactions to obstacles on road subjecting to given driving commands, i.e. follow, go straight, turn left and turn right for about 6 hours. A behavior-Cloning neural network architecture is proposed with the modified loss that enlarge the effects of errors for steer, which indicates the benefit to high an accuracy. We found the data augmentation of the image is crucial to the training of the proposed network. And a reasonable limit allows avoiding unexpected stop. The experiments demonstrate 3 obstacle avoidance cases: for the same type as the training dataset, other automobile and two-wheeled vehicles. Finally, the CARLA benchmark is also tested.

关键词： imitation learning obstacle avoidance end-to-end autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Low-rank subspace representation from optimal coded-aperture for unsupervised classification of hyperspectral imagery

arXiv

引用

arXiv 2020年

作者： Zhu, Jianchen Zhang, Tong Zhao, Shengjie Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai201804 China School of Software Engineering Tongji University Shanghai201804 China School of Electronic and Information Engineering Tongji University Shanghai201804 China

This paper aims at developing a clustering approach with spectral images directly from the compressive measurements of coded aperture snapshot spectral imager (CASSI). Assuming that compressed measurements often lie approximately in low dimensional subspaces corresponding to multiple classes, state of the art methods generally obtains optimal solution for each step separately but cannot guarantee that it will achieve the globally optimal clustering results. In this paper, a low-rank subspace representation (LRSR) algorithm is proposed to perform clustering on the compressed measurements. In addition, a subspace structured norm is added into the objective of low-rank representation problem exploiting the fact that each point in a union of subspaces can be expressed as a sparse linear combination of all other points and that the matrix of the points within each subspace is low rank. Simulation with real dataset illustrates the accuracy of the proposed spectral image clustering approach. Copyright © 2020, The Authors. All rights reserved.

关键词： Spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Distributed Computation Offloading using Deep Reinforcement Learning in Internet of Vehicles

Distributed Computation Offloading using Deep Reinforcement ...

引用

IEEE International Conference on Communications in China (ICCC)

作者： Chen Chen Zheng Wang Qingqi Pei Ci He Zhibin Dou State Key Laboratory of Integrated Services Networks Xidian University Xi'an P.R.China Key Laboratory of Industrial Internet of Things & Networked Control Ministry of Education Chongqing China The Key Laboratory of Embedded System and Service Computing (Tongji University) Ministry of Education Shanghai China The 54th Research Institute of China Electronics Technology Group Corporation Shijiazhuang China

ISBN: (数字)9781728173276

ISBN: (纸本)9781728173283

In this paper, we first take the moving vehicles as a RP (resource pool), by which we proposed a distributed computation offloading scheme to fully utilize the available resources and reduce task execution time in I0V (Internet of Vehicles). After that, we divide a complex task into many sub-tasks and indicate that how to assign these small tasks to satisfy the task execution time in RP is a NP problem. The executing time of a task is modeled as the longest calculation time among all small tasks, which is actually a min-max problem. For a dynamically vehicular environment, a distributed computing offloading strategy based on deep reinforcement learning is proposed to find the best offloading scheme to minimize the execution time of a task. Numerical results demonstrate that our scheme is better than Partial Flooding Algorithm and can make full use of the available computing resources of surrounding vehicles by considering the mobility of vehicles, the delay of communication transmission, and the separability of the tasks, thus greatly reducing the execution time of the computing tasks.

关键词： Computational modeling Simulation Reinforcement learning User experience Delays Internet Task analysis

来源：评论

学校读者我要写书评

暂无评论

Non-Intrusive Load Disaggregation Using Semi-Supervised Learning Method

Non-Intrusive Load Disaggregation Using Semi-Supervised Lear...

引用

International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)

作者： Nan Miao Shengjie Zhao Qingjiang Shi Rongqing Zhang The Key Laboratory of Embedded System and Service Computing of Ministry of Education Tongji University Shanghai China School of Software Engineering The Key Laboratory of Embedded System and Service Computing of Ministry of Education Tongji University Shanghai China School of Software Engineering Tongji University Shanghai China

ISBN: (数字)9781728159287

ISBN: (纸本)9781728159294

With the emerging of smart metering around the world, there is a growing demand to analyse the residential energy usage. In this paper, we propose a Deep Neural Network (DNN)-based approach for non-intrusive load monitoring (NILM), which can achieve effective and efficient estimation of individual appliance usage according to a single main meter reading in a non-intrusive manner. Considering practical situations, two training methods are provided. The first training approach is fully supervised learning, which requires a ground truth of label, indicating the state of the appliance (ON/OFF), to build a prediction model. The second training approach is semi-supervised learning, leading to better performance by F-Measure metric while only requiring some more unlabelled training data. Experimental results on the low-sample rate REDD dataset demonstrate the superior performance of our proposed DNN-based method compared with Hidden Markov Model (HMM)based and Graph Signal Processing (GSP)-based approaches.

关键词： Training Hidden Markov models Home appliances Neural networks Convolution Security Pattern analysis

来源：评论

学校读者我要写书评

暂无评论

Low-Rate Non-Intrusive Appliance Load Monitoring Based on Graph Signal Processing

Low-Rate Non-Intrusive Appliance Load Monitoring Based on Gr...

引用

International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)

作者： Bing Zhang Shengjie Zhao Qingjiang Shi Rongqing Zhang The Key Laboratory of Embedded System and Service Computing of Ministry of Education Tongji University Shanghai China School of Software Engineering The Key Laboratory of Embedded System and Service Computing of Ministry of Education Tongji University Shanghai China School of Software Engineering Tongji University Shanghai China

ISBN: (数字)9781728159287

ISBN: (纸本)9781728159294

Thanks to the large-scale smart meters deployments around the world, non-intrusive appliance load monitoring (NILM) is receiving popularity. It aims to disaggregate the total electricity load of a home into individual appliances without resorting to any specific appliance power monitors. NILM is worthy of broad attention owing to its facilitation in energy savings. This paper regards NILM as a classification task and proposes a two-step method based on graph signal processing (GSP). In the first step, a smoothest solution is obtained by minimizing the regularization term. In the second step, gradient projection method, which uses the obtained minimizer as a start point, is adopted to optimize the while objective function, where NILM is regarded as a constrained nonlinear programming problem. The experiment results based on the open-access data set REDD clearly demonstrate that the proposed GSP-based method achieves improved performance compared with other state-of-the-art low-rate NILM approaches.

关键词： domestic appliances gradient methods graph theory nonlinear programming power system measurement signal processing smart meters Smart meters Power system measurements signal processing Nonlinear Programming gradient methods Graph theory Appliances load monitoring

来源：评论

学校读者我要写书评

暂无评论

Traffic flow prediction model based on deep belief network and genetic algorithm

Traffic flow prediction model based on deep belief network a...

引用

作者： Zhang, Yaying Huang, Guan Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai200092 China

Traffic flow prediction plays an indispensable role in the intelligent transportation system. The effectiveness of traffic control and management relies heavily on the prediction accuracy. The authors propose a model based on deep belief networks (DBNs) to predict the traffic flow. Moreover, they use Fletcher-Reeves conjugate gradient algorithm to optimise the fine-tuning of model's parameters. Since the traffic flow has various features at different times such as weekday, weekend, daytime and nighttime, the hyper-parameters of the model should adapt to the time. Therefore, they employ the genetic algorithm to find the optimal hyper-parameters of DBN models for different times. The dataset from Caltrans Performance Measurement system was used to evaluate the performance of their models. The experimental results demonstrate that the proposed model achieved better performance in different times. © The Institution of Engineering and Technology 2018.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

A New One-Class Classification Method with Multiple Encoder-Decoder Pairs for Images

A New One-Class Classification Method with Multiple Encoder-...

引用

International Conference on Intelligent Human-Machine systems and Cybernetics, IHMSC

作者： Dongxiang Chen Chungang Yan Mimi Wang Department of Electronics and Information Engineering Tongji University The Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai China

One of the main destinations of image classification methods is to screen out the images belonging to the target class (positive) and identify the images of other classes (negative). Although most classifiers are trained on both positive samples and negative samples, in reality, negative samples are often unavailable. One-class classifiers trained only on positive samples, are proposed to solve this problem. However, how to train effectively a classifier remains a daunting challenge. Considering the success of deep learning in the field of computer vision in recent years, we propose a one-class classification model for images based on convolutional neural network (CNN). The model consists of two parts of networks with different responsibilities. The network of the first part works as the discriminator, used to identify whether the images are positive. The networks of the second part, each of which consists of an encoder-decoder pair work as the guiders of the discriminator. They guide the discriminator to learn what images it should identify to be negative. Different encoder-decoder pairs can restore images to different degrees. Images restored to different degrees can be used to train the discriminator. The discriminator learns that images under a specific restored degree don't belong to the target class. The experiment results on CIFAR-10 show that our method can achieve good performance, with less difficulty in training than the GAN-based counterparts'.

关键词：

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：