检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Wang, Yi-Bo Hang, Jun-Yi Zhang, Min-Ling School of Computer Science and Engineering Southeast University Nanjing210096 China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education China

In multi-label learning, each training instance is associated with multiple labels simultaneously. Traditional multi-label learning studies primarily focus on closed set scenario, i.e. the class label set of test data is identical to those used in training phase. Nevertheless, in numerous real-world scenarios, the environment is open and dynamic where unknown labels may emerge gradually during testing. In this paper, the problem of multi-label open set recognition (MLOSR) is investigated, which poses significant challenges in classifying and recognizing instances with unknown labels in multi-label setting. To enable open set multi-label prediction, a novel approach named SLAN is proposed by leveraging sub-labeling information enriched by structural information in the feature space. Accordingly, unknown labels are recognized by differentiating the sub-labeling information from holistic supervision. Experimental results on various datasets validate the effectiveness of the proposed approach in dealing with the MLOSR problem. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Binary Decomposition: A Problem Transformation Perspective for Open-Set Semi-Supervised Learning 41

Binary Decomposition: A Problem Transformation Perspective f...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Hang, Jun-Yi Zhang, Min-Ling School of Computer Science and Engineering Southeast University Nanjing210096 China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education China

Semi-supervised learning (SSL) is a classical machine learning paradigm dealing with labeled and unlabeled data. However, it often suffers performance degradation in real-world open-set scenarios, where unlabeled data contains outliers from novel categories that do not appear in labeled data. Existing studies commonly tackle this challenging open-set SSL problem with detect-and-filter strategy, which attempts to purify unlabeled data by detecting and filtering outliers. In this paper, we propose a novel binary decomposition strategy, which refrains from error-prone procedure of outlier detection by directly transforming the original open-set SSL problem into a number of standard binary SSL problems. Accordingly, a concise yet effective approach named BDMatch is presented. BDMatch confronts two attendant issues brought by binary decomposition, i.e. class-imbalance and representation-compromise, with adaptive logit adjustment and label-specific feature learning respectively. Comprehensive experiments on diversified benchmarks clearly validate the superiority of BDMatch as well as the effectiveness of our binary decomposition strategy. Copyright 2024 by the author(s)

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Weighted syncmer sampling improves long-read mapping

Weighted syncmer sampling improves long-read mapping

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Quan, Wei Kang, Jinjun Deng, Yanfei Liu, Zhuang Zhu, Xiao Quan, Guangri Yantai University School of Computer and Control Engineering Yantai China Xinjiang University of Political Science and Law School of Information and Network Security Tumxuk China

ISBN: (纸本)9798350386226

Syncmers represent a novel class of methods for selecting k-mers that exhibit robustness against mutations in flanking sequences and demonstrate superior conservation in mutated sequences. Nevertheless, syncmers may generate a higher frequency of repetitive seed matches compared to alternative techniques, such as minimizers, which can result in increased computational time. In this article, we introduce weighted minimizer sampling, which integrates weighted minimizer sampling and syncmer sampling to enhance the sensitivity and accuracy of long-read mapping. We modified two state-of-the-art long-read mappers, Minimap2 and Winnowmap, by substituting the sketching sampling methods with weighted minimizer sampling. We assessed their sensitivity and accuracy using simulated and real datasets. The experimental results indicate that weighted minimizer sampling significantly improves the sensitivity and accuracy of long-read mapping. The source code is available at GitHub: https://***/hexinkaoan/weighted-syncmer. © 2024 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Textual Data De-Privatization Scheme Based on Generative Adversarial networks 24th

Textual Data De-Privatization Scheme Based on Generative Adv...

引用

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Du, Yanning Xu, Jinnan Zhang, Yaling Wang, Yichuan Wang, Zhoukai School of Computer Science and Engineering Xi’an University of Technology Xi’an China Shaanxi Key Laboratory for Network Computing and Security Technology Xi’an China

ISBN: (纸本)9789819615506

In many fields, such as healthcare, finance, and scientific research, data sharing and collaboration are critical to achieving better outcomes. However, the sharing of personal data often involves privacy risks, so privacy-preserving techniques are needed to ensure data security and privacy. The superior performance and flexibility of generative models in data representation have led to significant progress in the development of data privacy. This paper proposes a textual data de-privatization scheme based on generative adversarial networks that well combines generative adversarial networks with privacy protection. The generated de-privacy data retains the statistical properties of the original data and effectively removes sensitive information. Moreover, feature sorting and reshaping modules are introduced to enable the generator to better capture the relationships between features, thus improving the quality of synthetic data. In this paper, the utility of synthetic data is evaluated from three aspects, including similarity assessment, privacy assessment, and model utility assessment. Experimental results show that this method achieves a good trade-off in terms of textual data privacy protection and data quality maintenance. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

MAPPO based Single-hop Task Offloading with Dynamic Vehicle Prediction in RSU-assisted IoV 10

MAPPO based Single-hop Task Offloading with Dynamic Vehicle ...

引用

10th IEEE Smart World Congress, SWC 2024

作者： Zhao, Wei Yang, Dongling Weng, Tangjie Xu, Xinwei Shao, Xun Liu, Zhi Anhui University of Technology School of Computer Science and Technology China Anhui University of Technology School of Management Science and Engineering China Toyohashi University of Technology Department of Electrical and Electronic Information Engineering Japan The University of Electro-Communications Department of Computer and Network Engineering Japan

ISBN: (纸本)9798331520861

In the realm of Internet of Vehicles (IoV), Roadside Units (RSUs) play a vital role due to their advanced sensing and computing capabilities. They support Mobile Edge Computing (MEC) by enabling real-time data processing and reducing latency and energy consumption. Given the large volume of data and the limited computing capacity of RSUs, tasks are often offloaded to other nodes, such as the cloud, other RSUs, or vehicles. However, due to the environment dynamic caused by vehicle mobility, the set of neighboring nodes for each RSU frequently changes, and variations in neighboring nodes are not perceived timely. In such cases, RSUs need to predict changes in neighboring nodes to ensure effective task offloading. Moreover, since each RSU can only perceive the surrounding environment, it is difficult for a single node to make optimal offloading decisions for each task independently. To address these issues, this paper proposes a distributed task offloading scheme that incorporates vehicle trajectory prediction. First, each RSU uses the vehicle trajectories predicted by the Gated Recurrent Unit (GRU) model to determine neighboring vehicles and update the set of neighboring nodes. Then, to enable each RSU to select a suitable computing node and minimize computation delay and energy consumption for the task, we adopt a distributed approach where each RSU independently makes offloading decisions based on local observations. To achieve this goal, we establish a 01 mathematical model and transform it into a Markov Decision Process (MDP). Subsequently, we propose a solution based on Multi-Agent Proximal Policy Optimization (MAPPO). Through extensive simulation experiments, we demonstrate that the scheme effectively reduces both long-term average processing delay and energy consumption. © 2024 IEEE.

关键词： MAPPO neighboring vehicle prediction RSU task offloading

来源：评论

学校读者我要写书评

暂无评论

Characterization and Properties of Bent-Negabent Functions

引用

Chinese Journal of Electronics 2022年第4期31卷 786-792页

作者： JIANG Niu ZHAO Min YANG Zhiyao ZHUO Zepeng CHEN Guolong School of Mathematical Sciences Huaibei Normal University Fujian Provincial Key Laboratory of Network Security and Cryptology Fujian Normal University School of Cyber Science University of Science and Technology of China School of Computer Engineering Bengbu University

A further characterization of the bentnegabent functions is presented. Based on the concept of complete mapping polynomial, we provide a necessary and sufficient condition for a class of quadratic Boolean functions to be bent-negabent. A new characterization of negabent functions can be described by using the parity of Hamming weight. We further generalize the classical convolution theorem and give the nega-Hadamard transform of the composition of a Boolean function and a vectorial Boolean function. The nega-Hadamard transform of a generalized indirect sum is calculated by this composition method.

关键词： Boolean function Bent-negabent function Walsh-Hadamard transform Nega-Hadamard transform

来源：评论

学校读者我要写书评

暂无评论

RIS-Assisted UAV-Enabled JSAC System: Joint Radio Resource and Phase Shift Optimization Framework

RIS-Assisted UAV-Enabled JSAC System: Joint Radio Resource a...

引用

2023 IEEE Globecom Workshops, GC Wkshps 2023

作者： Tun, Yan Kyaw Dán, György School of Electrical Engineering and Computer Science Kth Royal Institute of Technology Division of Network and Systems Engineering Stockholm Sweden

ISBN: (纸本)9798350370218

We propose a reconfigurable intelligent surface (RIS)-assisted unmanned aerial vehicle (UAV)-enabled joint communication and sensing (JSAC) system in which multiple JSAC nodes equipped with both sensing and communication functions are deployed to perform sensing and communication simultaneously, and a communication-enabled RIS is installed to provide strong wireless links for JSAC nodes to communicate with the UAV (i.e., aerial base station). We then formulate the problem of maximizing the sum communication rate of JSAC nodes while ensuring their minimum sensing performance requirements, i.e., minimum estimation rate requirements, by jointly optimizing sub-band assignment, adaptive bandwidth allocation, and the RIS phase shift. The resulting problem is a mixed-integer nonlinear programming (MILP) problem with a non-convex structure, thus, to solve the problem efficiently, we decompose the problem into two coupled sub-problems, a sub-band assignment and adaptive bandwidth allocation problem, and a RIS phase shift problem. We propose a low-complexity iterative algorithm for solving the sub-problems in an alternating manner. Extensive simulations show that the proposed algorithm can increase the communication rate by up to 27.51 %,48.747%, and 62.0942% compared to RSA, RPS, and RSRPS schemes. © 2023 IEEE.

关键词： dual decomposition Joint sensing and communication one-to-one matching game reconfigurable intelligent surface unmanned aerial vehicle whale optimization

来源：评论

学校读者我要写书评

暂无评论

Multi-hop Link Quality Prediction for DRL based Task Offloading in RSU-assisted IoV 10

Multi-hop Link Quality Prediction for DRL based Task Offload...

引用

10th IEEE Smart World Congress, SWC 2024

作者： Zhao, Wei Zuo, Qixue Weng, Tangjie Xu, Xinwei Shao, Xun Liu, Zhi Anhui University of Technology School of Computer Science and Technology China Anhui University of Technology School of Management Science and Engineering China Toyohashi University of Technology Department of Electrical and Electronic Information Engineering Japan The University of Electro-Communications Department of Computer and Network Engineering Japan

ISBN: (纸本)9798331520861

Roadside units (RSUs) with strong sensing abilities enhance the feasibility of the RSU-to-Everything (R2X) paradigm, providing crucial infrastructure support for mobile edge computing and reducing data processing latency. Due to the limited computing capacity of RSUs, it is beneficial to offload tasks to other RSUs, vehicles, and cloud servers for processing, allowing for the efficient utilization of computational resources. To leverage the abundant idle vehicle resources on the road, tasks can be offloaded to vehicles outside the coverage area of the RSUs through multi-hop transmission. Due to the high mobility of vehicles, there is a risk of communication link disconnection during task transmission, which poses a challenge to the reliable transmission of tasks. In this paper, we ensure the communication link quality among nodes based on the established vehicle mobility model, and design a mechanism for selecting forwarding vehicles. The goal is to find the path with the highest reliability between vehicles and the source RSU. To encourage nodes to participate in multi-hop transmission, a certain relay cost and computational cost are paid to the computing nodes. We establish a joint optimization model focusing on task delay and task cost. To achieve this, we develop a 0-1 mathematical model and then transform it into a Markov Decision Process (MDP). Building on this foundation, we propose a solution based on Deep Reinforcement Learning (DRL). Experimental results demonstrate that this method significantly reduces task delay and task cost. © 2024 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

Adversarial Robustness of Multi-agent Reinforcement Learning Secondary Control of Islanded Inverter-based AC Microgrids 14

Adversarial Robustness of Multi-agent Reinforcement Learning...

引用

14th IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids, SmartGridComm 2023

作者： Shereen, Ezzeldin Kazari, Kiarash Dan, Gyorgy Kth Royal Institute of Technology Division of Network and Systems Engineering School of Electrical Engineering and Computer Science Stockholm Sweden

ISBN: (纸本)9781665455541

Secondary control of voltage magnitude and frequency is essential to the stable and secure operation of microgrids (MGs). Recent years have witnessed an increasing interest in developing secondary controllers based on multi-agent reinforcement learning (MARL), in order to replace existing model-based controllers. Nonetheless, unlike the vulnerabilities of model-based controllers, the vulnerability of MARLbased MG secondary controllers has so far not been addressed. In this paper, we investigate the vulnerability of MARL controllers to false data injection attacks (FDIAs). Based on a formulation of MG secondary control as a partially observable stochastic game (POSG), we propose to formulate the problem of computing FDIAs as a partially observable Markov decision process (POMDP), and we use state-of-the-art RL algorithms for solving the resulting problem. Based on extensive simulations of a MG with 4 distributed generators (DGs), our results show that MARL-based secondary controllers are more resilient to FDIAs compared to state of the art model-based controllers, both in terms of attack impact and in terms of the effort needed for computing impactful attacks. Our results can serve as additional arguments for employing MARL in future MG control. © 2023 IEEE.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

A Semi-supervised Protein Complex Identification Algorithm Based on Sparseness Constraint

A Semi-supervised Protein Complex Identification Algorithm B...

引用

2024 International Conference on Applied Mathematics, Modelling and Statistics Application, AMMSA 2024

作者： Liu, Guangming Dang, Fengqing Wang, Bin Peng, Yonghong Wang, Xiaofan Tan, Li School of Computer Science and Engineering Shaanxi Key Laboratory for Network Computing and Security Technology Xi'An University of Technology Shaanxi Xi'an China Faculty of Science and Engineering Computing and Information Science Cambridge United Kingdom School of Computer and Information Engineering Beijing Technology and Business University Beijing China

Recognising protein complexes in protein interaction networks is crucial, but poses a major challenge due to the frequency of noisy interactions. These networks typically involve numerous protein complexes, with each protein generally only participating in a few complexes. Current recognition models often ignore this aspect. To address this problem, we present a semi-supervised protein complex identification algorithm that extends non-negative matrix factorization (NMF) with sparsity constraints. In contrast to conventional approaches that apply a global sparsity constraint to the entire matrix, our method imposes individual sparsity constraints on protein membership indicator vectors. This targeted strategy controls the algorithm more effectively. Our experimental results with yeast and human protein interaction networks show that our algorithm achieves higher accuracy in identifying protein complexes than leading contemporary methods. © Published under licence by IOP Publishing Ltd.

关键词： Non-negative matrix factorization

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：