Pub Date : 2023-09-22DOI: 10.26599/TST.2023.9010010
Jun Zhu;Hongyi Li;Tao Zhang
In recent years, Simultaneous Localization And Mapping (SLAM) technology has prevailed in a wide range of applications, such as autonomous driving, intelligent robots, Augmented Reality (AR), and Virtual Reality (VR). Multi-sensor fusion using the most popular three types of sensors (e.g., visual sensor, LiDAR sensor, and IMU) is becoming ubiquitous in SLAM, in part because of the complementary sensing capabilities and the inevitable shortages (e.g., low precision and long-term drift) of the stand-alone sensor in challenging environments. In this article, we survey thoroughly the research efforts taken in this field and strive to provide a concise but complete review of the related work. Firstly, a brief introduction of the state estimator formation in SLAM is presented. Secondly, the state-of-the-art algorithms of different multi-sensor fusion algorithms are given. Then we analyze the deficiencies associated with the reviewed approaches and formulate some future research considerations. This paper can be considered as a brief guide to newcomers and a comprehensive reference for experienced researchers and engineers to explore new interesting orientations.
{"title":"Camera, LiDAR, and IMU Based Multi-Sensor Fusion SLAM: A Survey","authors":"Jun Zhu;Hongyi Li;Tao Zhang","doi":"10.26599/TST.2023.9010010","DOIUrl":"https://doi.org/10.26599/TST.2023.9010010","url":null,"abstract":"In recent years, Simultaneous Localization And Mapping (SLAM) technology has prevailed in a wide range of applications, such as autonomous driving, intelligent robots, Augmented Reality (AR), and Virtual Reality (VR). Multi-sensor fusion using the most popular three types of sensors (e.g., visual sensor, LiDAR sensor, and IMU) is becoming ubiquitous in SLAM, in part because of the complementary sensing capabilities and the inevitable shortages (e.g., low precision and long-term drift) of the stand-alone sensor in challenging environments. In this article, we survey thoroughly the research efforts taken in this field and strive to provide a concise but complete review of the related work. Firstly, a brief introduction of the state estimator formation in SLAM is presented. Secondly, the state-of-the-art algorithms of different multi-sensor fusion algorithms are given. Then we analyze the deficiencies associated with the reviewed approaches and formulate some future research considerations. This paper can be considered as a brief guide to newcomers and a comprehensive reference for experienced researchers and engineers to explore new interesting orientations.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"415-429"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258154.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027595","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}
Time series clustering is a challenging problem due to the large-volume, high-dimensional, and warping characteristics of time series data. Traditional clustering methods often use a single criterion or distance measure, which may not capture all the features of the data. This paper proposes a novel method for time series clustering based on evolutionary multi-tasking optimization, termed i-MFEA, which uses an improved multifactorial evolutionary algorithm to optimize multiple clustering tasks simultaneously, each with a different validity index or distance measure. Therefore, i-MFEA can produce diverse and robust clustering solutions that satisfy various preferences of decision-makers. Experiments on two artificial datasets show that i-MFEA outperforms single-objective evolutionary algorithms and traditional clustering methods in terms of convergence speed and clustering quality. The paper also discusses how i-MFEA can address two long-standing issues in time series clustering: the choice of appropriate similarity measure and the number of clusters.
{"title":"Evolutionary Multi-Tasking Optimization for High-Efficiency Time Series Data Clustering","authors":"Rui Wang;Wenhua Li;Kaili Shen;Tao Zhang;Xiangke Liao","doi":"10.26599/TST.2023.9010036","DOIUrl":"https://doi.org/10.26599/TST.2023.9010036","url":null,"abstract":"Time series clustering is a challenging problem due to the large-volume, high-dimensional, and warping characteristics of time series data. Traditional clustering methods often use a single criterion or distance measure, which may not capture all the features of the data. This paper proposes a novel method for time series clustering based on evolutionary multi-tasking optimization, termed i-MFEA, which uses an improved multifactorial evolutionary algorithm to optimize multiple clustering tasks simultaneously, each with a different validity index or distance measure. Therefore, i-MFEA can produce diverse and robust clustering solutions that satisfy various preferences of decision-makers. Experiments on two artificial datasets show that i-MFEA outperforms single-objective evolutionary algorithms and traditional clustering methods in terms of convergence speed and clustering quality. The paper also discusses how i-MFEA can address two long-standing issues in time series clustering: the choice of appropriate similarity measure and the number of clusters.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"343-355"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258256.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027602","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}
Pub Date : 2023-09-22DOI: 10.26599/TST.2023.9010035
Xiaojie Dai;Xingxin Wang;Xue Qu;Baolei Mao;Wei Hu
Fault analysis is a frequently used side-channel attack for cryptanalysis. However, existing fault attack methods usually involve complex fault fusion analysis or computation-intensive statistical analysis of massive fault traces. In this work, we take a property-based formal verification approach to fault analysis. We derive fine-grained formal models for automatic fault propagation and fusion, which establish a mathematical foundation for precise measurement and formal reasoning of fault effects. We extract the correlations in fault effects in order to create properties for fault verification. We further propose a method for key recovery, by formally checking when the extracted properties can be satisfied with partial keys as the search variables. Experimental results using both unprotected and masked advanced encryption standard (AES) implementations show that our method has a key search complexity of 2 16 , which only requires two correct and faulty ciphertext pairs to determine the secret key, and does not assume knowledge about fault location or pattern.
{"title":"Fault Analysis on AES: A Property-Based Verification Perspective","authors":"Xiaojie Dai;Xingxin Wang;Xue Qu;Baolei Mao;Wei Hu","doi":"10.26599/TST.2023.9010035","DOIUrl":"https://doi.org/10.26599/TST.2023.9010035","url":null,"abstract":"Fault analysis is a frequently used side-channel attack for cryptanalysis. However, existing fault attack methods usually involve complex fault fusion analysis or computation-intensive statistical analysis of massive fault traces. In this work, we take a property-based formal verification approach to fault analysis. We derive fine-grained formal models for automatic fault propagation and fusion, which establish a mathematical foundation for precise measurement and formal reasoning of fault effects. We extract the correlations in fault effects in order to create properties for fault verification. We further propose a method for key recovery, by formally checking when the extracted properties can be satisfied with partial keys as the search variables. Experimental results using both unprotected and masked advanced encryption standard (AES) implementations show that our method has a key search complexity of 2\u0000<sup>16</sup>\u0000, which only requires two correct and faulty ciphertext pairs to determine the secret key, and does not assume knowledge about fault location or pattern.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"576-588"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258165.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027658","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}
Pub Date : 2023-09-22DOI: 10.26599/TST.2023.9010016
Xiaolin Chai;Yan Sun;Yan Gao
As one of the most popular visual programming languages, Scratch has a lot of evaluation around it. Reasonable evaluation can help programmers understand their projects better. At the same time, it can also provide a reference for them to browse other projects in the online community. Most of the existing evaluations on Scratch are carried from three perspectives: Computational Thinking (CT) ability, visual presentation aesthetics, and code quality. Among them, the assessment of CT and code quality is mainly carried out from the program script, while the evaluation of visual aesthetics is analyzed from the perspective of image sequences generated by project execution. The single-view evaluation focuses on the performance of a program in a certain aspect and is one-sided. In this paper, we propose a multi-view evaluation framework to integrate various evaluations using different policies. We quantitatively analyze the assessment of different views driven by data. Combined with overall evaluations that represent human opinions, we analyze their differences and connections. Through experiments, we determine the weights of different integration policies, the proposed multi-view evaluation method can generate evaluation results similar to human opinions.
{"title":"Towards Data-Driving Multi-View Evaluation Framework for Scratch","authors":"Xiaolin Chai;Yan Sun;Yan Gao","doi":"10.26599/TST.2023.9010016","DOIUrl":"https://doi.org/10.26599/TST.2023.9010016","url":null,"abstract":"As one of the most popular visual programming languages, Scratch has a lot of evaluation around it. Reasonable evaluation can help programmers understand their projects better. At the same time, it can also provide a reference for them to browse other projects in the online community. Most of the existing evaluations on Scratch are carried from three perspectives: Computational Thinking (CT) ability, visual presentation aesthetics, and code quality. Among them, the assessment of CT and code quality is mainly carried out from the program script, while the evaluation of visual aesthetics is analyzed from the perspective of image sequences generated by project execution. The single-view evaluation focuses on the performance of a program in a certain aspect and is one-sided. In this paper, we propose a multi-view evaluation framework to integrate various evaluations using different policies. We quantitatively analyze the assessment of different views driven by data. Combined with overall evaluations that represent human opinions, we analyze their differences and connections. Through experiments, we determine the weights of different integration policies, the proposed multi-view evaluation method can generate evaluation results similar to human opinions.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"517-528"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258248.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027660","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}
Pub Date : 2023-09-22DOI: 10.26599/TST.2023.9010071
Kainan Zhang;DongMyung Shin;Daehee Seo;Zhipeng Cai
Open-source licenses can promote the development of machine learning by allowing others to access, modify, and redistribute the training dataset. However, not all open-source licenses may be appropriate for data sharing, as some may not provide adequate protections for sensitive or personal information such as social network data. Additionally, some data may be subject to legal or regulatory restrictions that limit its sharing, regardless of the licensing model used. Hence, obtaining large amounts of labeled data can be difficult, time-consuming, or expensive in many real-world scenarios. Few-shot graph classification, as one application of meta-learning in supervised graph learning, aims to classify unseen graph types by only using a small amount of labeled data. However, the current graph neural network methods lack full usage of graph structures on molecular graphs and social network datasets. Since structural features are known to correlate with molecular properties in chemistry, structure information tends to be ignored with sufficient property information provided. Nevertheless, the common binary classification task of chemical compounds is unsuitable in the few-shot setting requiring novel labels. Hence, this paper focuses on the graph classification tasks of a social network, whose complex topology has an uncertain relationship with its nodes' attributes. With two multi-class graph datasets with large node-attribute dimensions constructed to facilitate the research, we propose a novel learning framework that integrates both meta-learning and contrastive learning to enhance the utilization of graph topological information. Extensive experiments demonstrate the competitive performance of our framework respective to other state-of-the-art methods.
{"title":"Few-Shot Graph Classification with Structural-Enhanced Contrastive Learning for Graph Data Copyright Protection","authors":"Kainan Zhang;DongMyung Shin;Daehee Seo;Zhipeng Cai","doi":"10.26599/TST.2023.9010071","DOIUrl":"https://doi.org/10.26599/TST.2023.9010071","url":null,"abstract":"Open-source licenses can promote the development of machine learning by allowing others to access, modify, and redistribute the training dataset. However, not all open-source licenses may be appropriate for data sharing, as some may not provide adequate protections for sensitive or personal information such as social network data. Additionally, some data may be subject to legal or regulatory restrictions that limit its sharing, regardless of the licensing model used. Hence, obtaining large amounts of labeled data can be difficult, time-consuming, or expensive in many real-world scenarios. Few-shot graph classification, as one application of meta-learning in supervised graph learning, aims to classify unseen graph types by only using a small amount of labeled data. However, the current graph neural network methods lack full usage of graph structures on molecular graphs and social network datasets. Since structural features are known to correlate with molecular properties in chemistry, structure information tends to be ignored with sufficient property information provided. Nevertheless, the common binary classification task of chemical compounds is unsuitable in the few-shot setting requiring novel labels. Hence, this paper focuses on the graph classification tasks of a social network, whose complex topology has an uncertain relationship with its nodes' attributes. With two multi-class graph datasets with large node-attribute dimensions constructed to facilitate the research, we propose a novel learning framework that integrates both meta-learning and contrastive learning to enhance the utilization of graph topological information. Extensive experiments demonstrate the competitive performance of our framework respective to other state-of-the-art methods.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"605-616"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258268.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68028923","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}
Pub Date : 2023-09-22DOI: 10.26599/TST.2023.9010007
Nasir Ahmad Jalali;Hongsong Chen
The widespread use of the Internet of Things (IoTs) and the rapid development of artificial intelligence technologies have enabled applications to cross commercial and industrial band settings. Within such systems, all participants related to commercial and industrial systems must communicate and generate data. However, due to the small storage capacities of IoT devices, they are required to store and transfer the generated data to third-party entity called “cloud”, which creates one single point to store their data. However, as the number of participants increases, the size of generated data also increases. Therefore, such a centralized mechanism for data collection and exchange between participants is likely to face numerous challenges in terms of security, privacy, and performance. To address these challenges, Federated Learning (FL) has been proposed as a reasonable decentralizing approach, in which clients no longer need to transfer and store real data in the central server. Instead, they only share updated training models that are trained over their private datasets. At the same time, FL enables clients in distributed systems to share their machine learning models collaboratively without their training data, thus reducing data privacy and security challeges. However, slow model training and the execution of additional unnecessary communication rounds may hinder FL applications from operating properly in a distributed system. Furthermore, these unnecessary communication rounds make the system vulnerable to security and privacy issues, because irrelevant model updates are sent between clients and servers. Thus, in this work, we propose an algorithm for fully homomorphic encryption called Cheon-Kim-Kim-Song (CKKS) to encrypt model parameters for their local information privacy-preserving function. The proposed solution uses the impetus term to speed up model convergence during the model training process. Furthermore, it establishes a secure communication channel between IoT devices and the server. We also use a lightweight secure transport protocol to mitigate the communication overhead, thereby improving communication security and efficiency with low communication latency between client and server.
物联网（IoT）的广泛使用和人工智能技术的快速发展使应用程序能够跨越商业和工业波段。在这样的系统中，所有与商业和工业系统相关的参与者都必须进行通信并生成数据。然而，由于物联网设备的存储容量较小，它们需要将生成的数据存储并传输到名为“云”的第三方实体，该实体创建一个存储数据的单点。然而，随着参与者数量的增加，生成的数据的大小也会增加。因此，这种集中的参与者之间的数据收集和交换机制可能会在安全性、隐私性和性能方面面临许多挑战。为了应对这些挑战，联合学习（FL）被认为是一种合理的去中心化方法，其中客户端不再需要在中央服务器中传输和存储真实数据。相反，他们只共享在私人数据集上训练的更新训练模型。同时，FL使分布式系统中的客户端能够在没有训练数据的情况下协作共享他们的机器学习模型，从而减少数据隐私和安全挑战。然而，缓慢的模型训练和额外的不必要的通信轮次的执行可能会阻碍FL应用程序在分布式系统中正常运行。此外，这些不必要的通信回合使系统容易受到安全和隐私问题的影响，因为不相关的模型更新是在客户端和服务器之间发送的。因此，在这项工作中，我们提出了一种称为Cheon Kim Kim Song（CKKS）的全同态加密算法，以加密模型参数的局部信息隐私保护函数。所提出的解决方案在模型训练过程中使用动力项来加速模型收敛。此外，它在物联网设备和服务器之间建立了一个安全的通信通道。我们还使用了一种轻量级的安全传输协议来减轻通信开销，从而提高了通信安全性和效率，降低了客户端和服务器之间的通信延迟。
{"title":"Federated Learning Security and Privacy-Preserving Algorithm and Experiments Research Under Internet of Things Critical Infrastructure","authors":"Nasir Ahmad Jalali;Hongsong Chen","doi":"10.26599/TST.2023.9010007","DOIUrl":"https://doi.org/10.26599/TST.2023.9010007","url":null,"abstract":"The widespread use of the Internet of Things (IoTs) and the rapid development of artificial intelligence technologies have enabled applications to cross commercial and industrial band settings. Within such systems, all participants related to commercial and industrial systems must communicate and generate data. However, due to the small storage capacities of IoT devices, they are required to store and transfer the generated data to third-party entity called “cloud”, which creates one single point to store their data. However, as the number of participants increases, the size of generated data also increases. Therefore, such a centralized mechanism for data collection and exchange between participants is likely to face numerous challenges in terms of security, privacy, and performance. To address these challenges, Federated Learning (FL) has been proposed as a reasonable decentralizing approach, in which clients no longer need to transfer and store real data in the central server. Instead, they only share updated training models that are trained over their private datasets. At the same time, FL enables clients in distributed systems to share their machine learning models collaboratively without their training data, thus reducing data privacy and security challeges. However, slow model training and the execution of additional unnecessary communication rounds may hinder FL applications from operating properly in a distributed system. Furthermore, these unnecessary communication rounds make the system vulnerable to security and privacy issues, because irrelevant model updates are sent between clients and servers. Thus, in this work, we propose an algorithm for fully homomorphic encryption called Cheon-Kim-Kim-Song (CKKS) to encrypt model parameters for their local information privacy-preserving function. The proposed solution uses the impetus term to speed up model convergence during the model training process. Furthermore, it establishes a secure communication channel between IoT devices and the server. We also use a lightweight secure transport protocol to mitigate the communication overhead, thereby improving communication security and efficiency with low communication latency between client and server.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"400-414"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258150.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027594","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}
Anti-spoofing is becoming a crucial technique for applications with high navigation accuracy and reliability requirements. Anti-spoofing technique based on Receiver Autonomous Integrity Monitoring (RAIM) is a good choice for most Global Navigation Satellite System (GNSS) receivers because it does not require any change to the hardware of the receiver. However, the conventional RAIM method can only detect and mitigate a single spoofing signal. Some improved RAIM methods can deal with more spoofing signals, but the computational complexity increases dramatically when the number of satellites in view increase or need additional information. This paper proposes a new RAIM method, called the SRV-RAIM method, which has a very low computation complexity regardless of the number of satellites in view and can deal with any number of spoofing signals. The key to the new method is the spatial distribution characteristic of the Satellites' Residual Vectors (SRV). In replay or generative spoofing scenarios, the pseudorange measurements of spoofing signals are consistent, the residual vectors of real satellites and those of spoofing satellites have good separation characteristics in spatial distribution. Based on this characteristic, the SRV-RAIM method is proposed, and the simulation results show that the method can separate the real signals and the spoofing signals with an average probability of 86.55% in the case of 12 visible satellites, regardless of the number of spoofing signals. Compared to the conventional traversal-RAIM method, the performance is only reduced by 3.59%, but the computational cost is reduced by 98.3%, so most of the GNSS receivers can run the SRV-RAIM algorithm in time.
{"title":"A GNSS Anti-Spoofing Technique Based on the Spatial Distribution Characteristic of the Residual Vectors","authors":"Qi'an Wu;Xiaowei Cui;Mingquan Lu;Pengxiang Yang;Peng Wu","doi":"10.26599/TST.2023.9010017","DOIUrl":"https://doi.org/10.26599/TST.2023.9010017","url":null,"abstract":"Anti-spoofing is becoming a crucial technique for applications with high navigation accuracy and reliability requirements. Anti-spoofing technique based on Receiver Autonomous Integrity Monitoring (RAIM) is a good choice for most Global Navigation Satellite System (GNSS) receivers because it does not require any change to the hardware of the receiver. However, the conventional RAIM method can only detect and mitigate a single spoofing signal. Some improved RAIM methods can deal with more spoofing signals, but the computational complexity increases dramatically when the number of satellites in view increase or need additional information. This paper proposes a new RAIM method, called the SRV-RAIM method, which has a very low computation complexity regardless of the number of satellites in view and can deal with any number of spoofing signals. The key to the new method is the spatial distribution characteristic of the Satellites' Residual Vectors (SRV). In replay or generative spoofing scenarios, the pseudorange measurements of spoofing signals are consistent, the residual vectors of real satellites and those of spoofing satellites have good separation characteristics in spatial distribution. Based on this characteristic, the SRV-RAIM method is proposed, and the simulation results show that the method can separate the real signals and the spoofing signals with an average probability of 86.55% in the case of 12 visible satellites, regardless of the number of spoofing signals. Compared to the conventional traversal-RAIM method, the performance is only reduced by 3.59%, but the computational cost is reduced by 98.3%, so most of the GNSS receivers can run the SRV-RAIM algorithm in time.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"457-468"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258250.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027596","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}
The on-demand food delivery (OFD) service has gained rapid development in the past decades but meanwhile encounters challenges for further improving operation quality. The order dispatching problem is one of the most concerning issues for the OFD platforms, which refer to dynamically dispatching a large number of orders to riders reasonably in very limited decision time. To solve such a challenging combinatorial optimization problem, an effective matching algorithm is proposed by fusing the reinforcement learning technique and the optimization method. First, to deal with the large-scale complexity, a decoupling method is designed by reducing the matching space between new orders and riders. Second, to overcome the high dynamism and satisfy the stringent requirements on decision time, a reinforcement learning based dispatching heuristic is presented. To be specific, a sequence-to-sequence neural network is constructed based on the problem characteristic to generate an order priority sequence. Besides, a training approach is specially designed to improve learning performance. Furthermore, a greedy heuristic is employed to effectively dispatch new orders according to the order priority sequence. On real-world datasets, numerical experiments are conducted to validate the effectiveness of the proposed algorithm. Statistical results show that the proposed algorithm can effectively solve the problem by improving delivery efficiency and maintaining customer satisfaction.
{"title":"A Matching Algorithm with Reinforcement Learning and Decoupling Strategy for Order Dispatching in On-Demand Food Delivery","authors":"Jingfang Chen;Ling Wang;Zixiao Pan;Yuting Wu;Jie Zheng;Xuetao Ding","doi":"10.26599/TST.2023.9010069","DOIUrl":"https://doi.org/10.26599/TST.2023.9010069","url":null,"abstract":"The on-demand food delivery (OFD) service has gained rapid development in the past decades but meanwhile encounters challenges for further improving operation quality. The order dispatching problem is one of the most concerning issues for the OFD platforms, which refer to dynamically dispatching a large number of orders to riders reasonably in very limited decision time. To solve such a challenging combinatorial optimization problem, an effective matching algorithm is proposed by fusing the reinforcement learning technique and the optimization method. First, to deal with the large-scale complexity, a decoupling method is designed by reducing the matching space between new orders and riders. Second, to overcome the high dynamism and satisfy the stringent requirements on decision time, a reinforcement learning based dispatching heuristic is presented. To be specific, a sequence-to-sequence neural network is constructed based on the problem characteristic to generate an order priority sequence. Besides, a training approach is specially designed to improve learning performance. Furthermore, a greedy heuristic is employed to effectively dispatch new orders according to the order priority sequence. On real-world datasets, numerical experiments are conducted to validate the effectiveness of the proposed algorithm. Statistical results show that the proposed algorithm can effectively solve the problem by improving delivery efficiency and maintaining customer satisfaction.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"386-399"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258151.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027600","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}
Pub Date : 2023-09-22DOI: 10.26599/TST.2023.9010023
Yuan Zhou;Peng Wang;Lei Xiang;Haofeng Zhang
Recently, Generative Adversarial Networks (GANs) have become the mainstream text-to-image (T2I) framework. However, a standard normal distribution noise of inputs cannot provide sufficient information to synthesize an image that approaches the ground-truth image distribution. Moreover, the multistage generation strategy results in complex T2I applications. Therefore, this study proposes a novel feature-grounded single-stage T2I model, which considers the “real” distribution learned from training images as one input and introduces a worst-case-optimized similarity measure into the loss function to enhance the model's generation capacity. Experimental results on two benchmark datasets demonstrate the competitive performance of the proposed model in terms of the Frechet inception distance and inception score compared to those of some classical and state-of-the-art models, showing the improved similarities among the generated image, text, and ground truth.
{"title":"Feature-Grounded Single-Stage Text-to-Image Generation","authors":"Yuan Zhou;Peng Wang;Lei Xiang;Haofeng Zhang","doi":"10.26599/TST.2023.9010023","DOIUrl":"https://doi.org/10.26599/TST.2023.9010023","url":null,"abstract":"Recently, Generative Adversarial Networks (GANs) have become the mainstream text-to-image (T2I) framework. However, a standard normal distribution noise of inputs cannot provide sufficient information to synthesize an image that approaches the ground-truth image distribution. Moreover, the multistage generation strategy results in complex T2I applications. Therefore, this study proposes a novel feature-grounded single-stage T2I model, which considers the “real” distribution learned from training images as one input and introduces a worst-case-optimized similarity measure into the loss function to enhance the model's generation capacity. Experimental results on two benchmark datasets demonstrate the competitive performance of the proposed model in terms of the Frechet inception distance and inception score compared to those of some classical and state-of-the-art models, showing the improved similarities among the generated image, text, and ground truth.","PeriodicalId":60306,"journal":{"name":"Tsinghua Science and Technology","volume":"29 2","pages":"469-480"},"PeriodicalIF":6.6,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/iel7/5971803/10258149/10258251.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"68027435","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：