Tsinghua Science and Technology最新文献_第10页

Maximization of k-Submodular Function with d-Knapsack Constraints Over Sliding Window 具有 d-Knapsack 约束条件的 k 次模态函数在滑动窗口上的最大化

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2023.9010121

Wenqi Wang;Yuefang Sun;Zhiren Sun;Donglei Du;Xiaoyan Zhang

引用次数: 0

Key Mechanisms on Resource Optimization Allocation in Minority Game Based on Reinforcement Learning 基于强化学习的少数派博弈资源优化配置关键机制研究

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2023.9010155

Changyan Di;Tianyi Wang;Qingguo Zhou;Jinqiang Wang

{"title":"Key Mechanisms on Resource Optimization Allocation in Minority Game Based on Reinforcement Learning","authors":"Changyan Di;Tianyi Wang;Qingguo Zhou;Jinqiang Wang","doi":"10.26599/TST.2023.9010155","DOIUrl":"https://doi.org/10.26599/TST.2023.9010155","url":null,"abstract":"The emergence of coordinated and consistent macro behavior among self-interested individuals competing for limited resources represents a central inquiry in comprehending market mechanisms and collective behavior. Traditional economics tackles this challenge through a mathematical and theoretical lens, assuming individuals are entirely rational and markets tend to stabilize through the price mechanism. Our paper addresses this issue from an econophysics standpoint, employing reinforcement learning to construct a multi-agent system modeled on minority games. Our study has undertaken a comparative analysis from both collective and individual perspectives, affirming the pivotal roles of reward feedback and individual memory in addressing the aforementioned challenge. Reward feedback serves as the guiding force for the evolution of collective behavior, propelling it towards an overall increase in rewards. Individuals, drawing insights from their own rewards through accumulated learning, gain information about the collective state and adjust their behavior accordingly. Furthermore, we apply information theory to present a formalized equation for the evolution of collective behavior. Our research supplements existing conclusions regarding the mechanisms of a free market and, at a micro level, unveils the dynamic evolution of individual behavior in synchronization with the collective.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"721-731"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786935","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Study of Driver's Perception in Driving Tasks Based on Naturalistic Driving Experiments and fNIRS Measurement 基于自然驾驶实验和 fNIRS 测量的驾驶员驾驶任务感知研究

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2024.9010002

Bilu Li;Xin Pei;Dan Zhang;Xinmiao Zhang;Zhuoran Li;Duanrui Yu;Shifei Shen

引用次数: 0

Attention-Enhanced and Knowledge-Fused Dual Item Representations Network for Recommendation 基于注意增强和知识融合的双项目推荐表示网络

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2023.9010143

Qiang Hua;Jiachao Zhou;Feng Zhang;Chunru Dong;Dachuan Xu

{"title":"Attention-Enhanced and Knowledge-Fused Dual Item Representations Network for Recommendation","authors":"Qiang Hua;Jiachao Zhou;Feng Zhang;Chunru Dong;Dachuan Xu","doi":"10.26599/TST.2023.9010143","DOIUrl":"https://doi.org/10.26599/TST.2023.9010143","url":null,"abstract":"Integrating Knowledge Graphs (KGs) into recommendation systems as supplementary information has become a prevalent strategy. By leveraging the semantic relationships between entities in KGs, recommendation systems can better comprehend user preferences. Due to the unique structure of KGs, methods based on Graph Neural Networks (GNNs) have emerged as the current technical trend. However, existing GNN-based methods struggle to (1) filter out noisy information in real-world KGs, and (2) differentiate the item representations obtained from the knowledge graph and bipartite graph. In this paper, we introduce a novel model called Attention-enhanced and Knowledge-fused Dual item representations Network for recommendation (namely AKDN) that employs attention and gated mechanisms to guide aggregation on both knowledge graphs and bipartite graphs. In particular, we firstly design an attention mechanism to determine the weight of each edge in the information aggregation on KGs, which reduces the influence of noisy information on the items and enables us to obtain more accurate and robust representations of the items. Furthermore, we exploit a gated aggregation mechanism to differentiate collaborative signals and knowledge information, and leverage dual item representations to fuse them together for better capturing user behavior patterns. We conduct extensive experiments on two public datasets which demonstrate the superior performance of our AKDN over state-of-the-art methods, like Knowledge Graph Attention Network (KGAT) and Knowledge Graph- based Intent Network (KGIN).","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"585-599"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786928","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ILP-Based Heuristics for the Multi-Modal Stable Matching Problem 基于ilp的多模态稳定匹配问题的启发式算法

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2023.9010135

Yang Yang;Rolf H. Möhring;Junteng Song;Yicheng Xu;Yong Zhang

{"title":"ILP-Based Heuristics for the Multi-Modal Stable Matching Problem","authors":"Yang Yang;Rolf H. Möhring;Junteng Song;Yicheng Xu;Yong Zhang","doi":"10.26599/TST.2023.9010135","DOIUrl":"https://doi.org/10.26599/TST.2023.9010135","url":null,"abstract":"In this paper, we investigate the stable matching problem with multiple preferences in bipartite graphs, where each agent has various preference lists for all available partners with respect to different criteria. The problem requires that each matched agent must have exactly one partner and the obtained matching should be stable for all criteria. As our main contribution, we present an integer linear programming (ILP) model for determining whether there exists a globally stable matching in bipartite graphs, which has been proved to be NP-hard. Since the time consumed for solving ILPs might dramatically increase as the size of instances grows, we develop a preprocessing technique that helps to eliminate pairs that will never be a member of any globally stable matching and thus accelerates the computing process. We perform experiments on randomly generated preference lists and observe a significant speedup when we preprocess the instance before solving the ILPs. As there does not need to exist a perfect matching that is stable for all given criteria, we extend our ILP to the optimized version of the aforementioned problem, which asks to find a matching with maximum cardinality that is stable among all matched agents.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"479-487"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786943","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Fall Detection Device Based on Single Sensor Combined with Joint Features 基于单传感器和关节特征的跌倒检测设备

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2023.9010129

Li Zhang;Yu-An Liu;Qiuyu Wang;Huilin Chen;Jingao Xu;Danyang Li

{"title":"A Fall Detection Device Based on Single Sensor Combined with Joint Features","authors":"Li Zhang;Yu-An Liu;Qiuyu Wang;Huilin Chen;Jingao Xu;Danyang Li","doi":"10.26599/TST.2023.9010129","DOIUrl":"https://doi.org/10.26599/TST.2023.9010129","url":null,"abstract":"Accidental falls pose a significant threat to the well-being of the elderly, thus facilitating a quantum leap in the field of fall detection technology. For fall detection, accurate identification of fall behavior is a key priority. Our study proposes an innovative methodology to detect falls during activities of daily living (ADL), with the objective of preventing further harm. Our design aims to achieve precise identification of falls by extracting a variety of features obtained from the simultaneous acquisition of acceleration and angular velocity data using a single sensor. To enhance detection accuracy and reduce false alarms, we establish a classifier based on the joint acceleration and Euler angle feature (JAEF) analysis. With the aid of a support vector machine (SVM) classifier, human activities are classified into eight categories: going upstairs, going downstairs, running, walking, falling forward, falling backward, falling left, and falling right. In particular, we introduce a novel approach to enhance the accuracy of fall detection algorithms by introducing the Equal Signal Amplitude Difference method. Through experimental demonstration, the proposed method exhibits a remarkable sensitivity of 99.25%, precision of 98.75%, and excels in classification accuracy. It is noteworthy that the utilization of multiple features proves more effective than relying solely on a single aspect. The preliminary findings highlight the promising applications of our study in the field of fall injury systems.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"695-707"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786934","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Traceability-Based Operation with Retailer's Altruistic Preference Under Blockchain Technology 区块链技术下基于零售商利他主义偏好的溯源运营

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2023.9010120

Chunming Xu;Yi Zhao;Chenchen Wu

{"title":"A Traceability-Based Operation with Retailer's Altruistic Preference Under Blockchain Technology","authors":"Chunming Xu;Yi Zhao;Chenchen Wu","doi":"10.26599/TST.2023.9010120","DOIUrl":"https://doi.org/10.26599/TST.2023.9010120","url":null,"abstract":"The application of Blockchain Technology (BT) makes consumers trace more product information and enhances their trust in the product brand, but it also brings cost pressure to some participants who adopt the technology. Thus, it is particularly important how to encourage and support more enterprises to participate in BT. This study incorporates the altruistic preference into a BT-enabled three-echelon supply chain consisting of a powerful retailer, a supplier, and a manufacturer. We investigate the optimal strategies of the supply chain in different scenarios: without and with the BT, and without and with the altruistic preference. Then, we explore the effects of the BT and the retailer's altruistic preference on the profit performance of the supply chain system. The results show that the retailer, the manufacturer, and the supplier do not always benefit from the traceability of the BT, and while the powerful retailer's altruistic preference decreases his or her own profit but increases the profits of the manufacturer, the supplier, and the entire supply chain. Finally, a profit redistribution mechanism is designed to coordinate the supply chain with the retailer's altruistic preference under the BT.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"499-518"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786951","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cooperative Digital Healthcare Task Scheduling and Resource Management in Edge Intelligence Systems 边缘智能系统中的协同数字医疗任务调度和资源管理

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2024.9010140

Xing Liu;Jianhui Lv;Byung-Gyu Kim;Keqin Li;Hongkai Jin;Wei Gao;Jiayuan Bai

{"title":"Cooperative Digital Healthcare Task Scheduling and Resource Management in Edge Intelligence Systems","authors":"Xing Liu;Jianhui Lv;Byung-Gyu Kim;Keqin Li;Hongkai Jin;Wei Gao;Jiayuan Bai","doi":"10.26599/TST.2024.9010140","DOIUrl":"https://doi.org/10.26599/TST.2024.9010140","url":null,"abstract":"The rapid growth of digital healthcare applications has led to an increasing demand for efficient and reliable task scheduling and resource management in edge computing environments. However, the limited resources of edge servers and the need to process delay-sensitive healthcare tasks pose significant challenges. Existing solutions often need help to balance the trade-off between system cost and quality of service, particularly in resource-constrained scenarios. To address these challenges, we propose a novel cooperative task scheduling and resource management framework for digital healthcare applications in edge intelligence systems. Our approach leverages a two-step optimization strategy that combines the Multi-armed Combinatorial Selection Problem (MCSP) for task scheduling and the Sequential Markov Decision Process (SMDP) with alternative reward estimation for computation offloading. The MCSP-based scheduling algorithm efficiently explores the combinatorial task scheduling space to minimize healthcare task completion time and costs. The SMDP-based offloading strategy incorporates alternative reward estimation to improve robustness against dynamic variations in the system environment. Extensive simulations using real-world healthcare data demonstrate the superior performance of our proposed framework compared to state-of-the-art baselines, achieving significant improvements in cost, task success rate, and fairness. The proposed approach enables reliable and efficient digital healthcare services in resource-constrained edge computing environments.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"926-945"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786949","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142797888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Multi-Hyperparameter Prediction Framework for Distributed Energy Trading on Photovoltaic Network 光伏网络分布式能源交易的多参数预测框架

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2024.9010150

Chun Chen;Yong Zhang;Boon Han Lim;Li Ning;Shengzhong Feng;Peng Xie

{"title":"A Multi-Hyperparameter Prediction Framework for Distributed Energy Trading on Photovoltaic Network","authors":"Chun Chen;Yong Zhang;Boon Han Lim;Li Ning;Shengzhong Feng;Peng Xie","doi":"10.26599/TST.2024.9010150","DOIUrl":"https://doi.org/10.26599/TST.2024.9010150","url":null,"abstract":"The rapid evolution of distributed energy resources, particularly photovoltaic systems, poses a formidable challenge in maintaining a delicate balance between energy supply and demand while minimizing costs. The integrated nature of distributed markets, blending centralized and decentralized elements, holds the promise of maximizing social welfare and significantly reducing overall costs, including computational and communication expenses. However, achieving this balance requires careful consideration of various hyperparameter sets, encompassing factors such as the number of communities, community detection methods, and trading mechanisms employed among nodes. To address this challenge, we introduce a groundbreaking neural network-based framework, the Energy Trading-based Artificial Neural Network (ET-ANN), which excels in performance compared to existing algorithms. Our experiments underscore the superiority of ET-ANN in minimizing total energy transaction costs while maximizing social welfare within the realm of photovoltaic networks.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"864-874"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786948","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142797955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Machine Learning Rapid Prediction of the Aerothermodynamic Environment for Near-Space Hypersonic Unmanned Aircraft 近空间高超音速无人驾驶飞机空气热动力环境的机器学习快速预测

IF 6.6 1区计算机科学

Tsinghua Science and Technology Pub Date : 2024-12-09 DOI: 10.26599/TST.2024.9010018

Xujia Chen;Wenhui Fan

{"title":"A Machine Learning Rapid Prediction of the Aerothermodynamic Environment for Near-Space Hypersonic Unmanned Aircraft","authors":"Xujia Chen;Wenhui Fan","doi":"10.26599/TST.2024.9010018","DOIUrl":"https://doi.org/10.26599/TST.2024.9010018","url":null,"abstract":"Near-space hypersonic unmanned aircrafts (NHUA) encounter significant aerodynamic heating effects when flying at high velocities in extreme conditions. This leads to the generation of extremely high temperatures, reaching several thousand degrees, posing a substantial risk to the safety of NHUA. Accurate and rapid prediction of the aerothermodynamic environment is crucial for the thermal protection of NHUA. Conventional approaches exhibit some limitations, including the need for extensive pre-processing, long calculation time, inadequate precision, and reliance on expert knowledge, making them ill-suited for online intelligent prediction. This study proposes a novel “flying state-pressure and heat flux-temperature” data-driven prediction theoretical framework, considering both efficiency and accuracy. Our approach entails a prediction model for high-dimensional pressure and heat flux fields, employing principal component analysis (PCA) and multi-layer perceptron (MLP) models. A temperature time series model is also constructed using recurrent neural networks (RNN). The experimental results suggest that the prediction error falls within a narrow margin of approximately 5%. It takes around 0.1 seconds to forecast a high-dimensional field and 1 second to predict the temperature time series, which satisfies both speed and accuracy requirements.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 2","pages":"682-694"},"PeriodicalIF":6.6,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10786937","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825975","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0