检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Zhang, Xinzhi Zhu, Qingling Lin, Qiuzhen Chen, Wei-Neng Li, Jianqiang Coello, Carlos Artemio Coello College of Computer Science and Software Engineering Shenzhen University Shenzhen China School of Computers South China University of Technology Guangzhou510006 China Department of Computer Science MexicoD.F. 07360 Mexico Ikerbasque Spain

The Nurse Rostering Problem (NRP) is important for hospital management, which aims to balance both the needs of employees and the requirements of hospital operations. Currently, most studies often use local search methods with adaptive neighborhood search (ANS) or variable neighborhood descent (VND) to efficiently tackle the NRP. However, the use of ANS or VND in their local search strategies typically focuses on optimizing the current solution without considering the potential long-term consequences of these local improvements, which may easily fall into local optima and resultantly limit the final performance. Thus, this paper proposes a Deep Q-network-based Neighborhood Tabu Search (called DQN-NTS) method for tackling the NRP. Our method is a first attempt to train a deep Q-network (DQN) as a selection method for identifying effective sequences of neighborhoods, which can learn good transitions among the neighborhoods. We model the neighborhood search as a Markov model, where the current solution represents the state, and the reward signifies the improvement in the objective function. This approach enables DQN to consider neighborhood transition sequences that have historically obtained superior solutions, thereby helping to avoid local optima and achieve global optima by considering long-term impacts. Moreover, tabu search is combined with the DQN to iterate better solutions and diversify the search space by changing the weights of nurses when stagnating at the local optimum. Experimental results indicate that DQN-NTS outperforms other state-of-the-art tabu search methods with ANS and VND when running on the Second International Nurse Rostering Competition dataset. © 2023, The Authors. All rights reserved.

关键词： Tabu search

来源：评论

学校读者我要写书评

暂无评论

Indoor Drone Localization and Tracking Based on Acoustic Inertial Measurement

arXiv

引用

arXiv 2025年

作者： Sun, Yimiao Wang, Weiguo Mottola, Luca Jia, Zhang Wang, Ruijin He, Yuan School of Software BNRist Tsinghua University Beijing100190 China Politecnico di Milano Milano20133 Italy RI.SE Sweden Uppsala University Uppsala751 05 Sweden School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu610056 China

We present Acoustic Inertial Measurement (AIM), a one-of-a-kind technique for indoor drone localization and tracking. Indoor drone localization and tracking are arguably a crucial, yet unsolved challenge: in GPS-denied environments, existing approaches enjoy limited applicability, especially in Non-Line of Sight (NLoS), require extensive environment instrumentation, or demand considerable hardware/software changes on drones. In contrast, AIM exploits the acoustic characteristics of the drones to estimate their location and derive their motion, even in NLoS settings. We tame location estimation errors using a dedicated Kalman filter and the Interquartile Range rule (IQR) and demonstrate that AIM can support indoor spaces with arbitrary ranges and layouts. We implement AIM using an off-the-shelf microphone array and evaluate its performance with a commercial drone under varied settings. Results indicate that the mean localization error of AIM is 46% lower than that of commercial UWB-based systems in a complex 10 m×10 m indoor scenario, where state-of-the-art infrared systems would not even work because of NLoS situations. When distributed microphone arrays are deployed, the mean error can be reduced to less than 0.5m in a 20m range, and even support spaces with arbitrary ranges and layouts. Copyright © 2025, The Authors. All rights reserved.

关键词： Acoustic wave velocity measurement

来源：评论

学校读者我要写书评

暂无评论

Algorithm Modeling of Smile Intensity Estimation Using a Spatial Attention Convolutional Neural Network 3

Algorithm Modeling of Smile Intensity Estimation Using a Spa...

引用

2020 3rd International Conference on Modeling, Simulation and Optimization Technologies and Applications, MSOTA 2020

作者： Luo, Zhenzhen Sha, Zongyu Jin, Chen Xu, Ning School of Software Jiangxi Normal University Jiangxi China School of Computer Science and Technology Wuhan University of Technology Wuhan China School of Information Engineering Wuhan University of Technology Wuhan China

Smile intensity estimation is a challenging task as it required subtle feature extraction, self-Adapted weighted model and classifier. complexity of the problem domains, and problems on fine-grained image recognition are some of the issues related to intensity estimation. In this study, we designed a self-weighted deep convolutional neural network architecture for smiles intensity estimation using graphics processing unit. In the case of using only CK+ smile images, the accuracy of the model is also higher than that of the latest technology. Our model achieved better accuracy by just using CK+ smile images than state-of-The-Art techniques. Visualizations of learned features at various layers and their deconvolutions are also presented for understanding the learning process. © Published under licence by IOP Publishing Ltd.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Federated Learning with Client Availability Budgets

Federated Learning with Client Availability Budgets

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Yunkai Bao Steve Drew Xin Wang Jiayu Zhou Xiaoguang Niu Department of Electrical and Software Engineering University of Calgary Calgary AB Canada Department of Geomatics Engineering University of Calgary Calgary AB Canada Department of Computer Science and Engineering Michigan State University East Lansing MI USA School of Computer Science Wuhan University Wuhan Hubei China

Federated learning (FL) sheds light on efficiently and privately learning from massive Internet of Things (IoT) devices. However, the iterative training and aggregation pose additional stress on the limited energy and availability budgets of clients. In this paper, we discuss two types of availability budgets of IoT clients, including the timing to start participating in FL and the communication budgets due to their constrained energy. We theoretically analyze the effect of availability budgets on FL, based on the availability constraints, by leveraging a decaying quadratic function to prioritize learning from statistically heterogeneous clients during the initial training rounds. We also consider the effects of client availability in terms of their participation to find a balance among clients with varying availability. We present FedCAB, an algorithm applying our theoretical model for the probabilistic rankings of the available clients to select in each round of FL model aggregation. Numerical results show the effectiveness of FedCAB under label distribution skew with a limited communication budget and clients that join the learning process in later rounds. We release the source code of FedCAB at https://***/denoslab/FedCAB.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Perceptual Audio Object Coding Using Adaptive Subband Grouping with CNN and Residual Block

Perceptual Audio Object Coding Using Adaptive Subband Groupi...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yulin Wu Ruimin Hu Xiaochen Wang National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China

Spatial audio content is becoming increasingly popular and is regarded as a set of object signals with associated metadata. The object-based content representation is independent of loudspeaker layouts and provides high spatial resolution when reproduced on more loudspeakers. The audio quality of the traditional spatial audio object coding (SAOC) method has severe aliasing distortion, which impairs the immersive listening experience. In this study, we reduce aliasing distortion by perceptual adaptive subband grouping strategy and use the convolutional neural network (CNN) and residual block to build the side information compressing model. Both objective and subjective experiments on benchmark datasets with different bitrates show that the proposed method achieves favorable performance against state-of-the-art methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FireXnet: an explainable AI-based tailored deep learning model for wildfire detection on resource-constrained devices

引用

Fire Ecology 2023年第1期19卷 1-19页

作者： Ahmad, Khubab Khan, Muhammad Shahbaz Ahmed, Fawad Driss, Maha Boulila, Wadii Alazeb, Abdulwahab Alsulami, Mohammad Alshehri, Mohammed S. Ghadi, Yazeed Yasin Ahmad, Jawad Faculty of Engineering and Technology Multimedia University Melaka Malaysia School of Computing Engineering and the Built Environment Edinburgh Napier University Edinburgh UK Department of Cyber Security Pakistan Navy Engineering College NUST Karachi Pakistan Robotics and Internet of Things Laboratory Prince Sultan University Riyadh Saudi Arabia RIADI Laboratory National School of Computer Sciences University of Manouba Manouba Tunisia Department of Computer Science College of Computer Science and Information Systems Najran University Najran Saudi Arabia Department of Software Engineering Computer Science Al Ain University Abu Dhabi United Arab Emirates

Forests cover nearly one-third of the Earth’s land and are some of our most biodiverse ecosystems. Due to climate change, these essential habitats are endangered by increasing wildfires. Wildfires are not just a risk to the environment, but they also pose public health risks. Given these issues, there is an indispensable need for efficient and early detection methods. Conventional detection approaches fall short due to spatial limitations and manual feature engineering, which calls for the exploration and development of data-driven deep learning solutions. This paper, in this regard, proposes 'FireXnet', a tailored deep learning model designed for improved efficiency and accuracy in wildfire detection. FireXnet is tailored to have a lightweight architecture that exhibits high accuracy with significantly less training and testing time. It contains considerably reduced trainable and non-trainable parameters, which makes it suitable for resource-constrained devices. To make the FireXnet model visually explainable and trustable, a powerful explainable artificial intelligence (AI) tool, SHAP (SHapley Additive exPlanations) has been incorporated. It interprets FireXnet’s decisions by computing the contribution of each feature to the prediction. Furthermore, the performance of FireXnet is compared against five pre-trained models — VGG16, InceptionResNetV2, InceptionV3, DenseNet201, and MobileNetV2 — to benchmark its efficiency. For a fair comparison, transfer learning and fine-tuning have been applied to the aforementioned models to retrain the models on our dataset. The test accuracy of the proposed FireXnet model is 98.42%, which is greater than all other models used for comparison. Furthermore, results of reliability parameters confirm the model’s reliability, i.e., a confidence interval of [0.97, 1.00] validates the certainty of the proposed model’s estimates and a Cohen’s kappa coefficient of 0.98 proves that decisions of FireXnet are in considerable accordance with t

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generalizing Few Data to Unseen Domains Flexibly Based on Label Smoothing Integrated with Distributionally Robust Optimization

arXiv

引用

arXiv 2024年

作者： Wang, Yangdi Zhang, Zhi-Hai Xu, Su Xiu Guo, Wenming School of Management Beijing Institute of Technology Beijing100081 China Department of Industrial Engineering Tsinghua University Beijing100084 China School of Management Beijing Institute of Technology Beijing100081 China School of Computer Science National Pilot Software Engineering School Beijing University of Posts and Telecommunications Beijing100876 China

Overfitting commonly occurs when applying deep neural networks (DNNs) on small-scale datasets, where DNNs do not generalize well from existing data to unseen data. The main reason resulting in overfitting is that small-scale datasets cannot reflect the situations of the real world. Label smoothing (LS) is an effective regularization method to prevent overfitting, avoiding it by mixing one-hot labels with uniform label vectors. However, LS only focuses on labels while ignoring the distribution of existing data. In this paper, we introduce the distributionally robust optimization (DRO) to LS, achieving shift the existing data distribution flexibly to unseen domains when training DNNs. Specifically, we prove that the regularization of LS can be extended to a regularization term for the DNNs parameters when integrating DRO. The regularization term can be utilized to shift existing data to unseen domains and generate new data. Furthermore, we propose an approximate gradient-iteration label smoothing algorithm (GI-LS) to achieve the findings and train DNNs. We prove that the shift for the existing data does not influence the convergence of GI-LS. Since GI-LS incorporates a series of hyperparameters, we further consider using Bayesian optimization (BO) to find the relatively optimal combinations of these hyperparameters. Taking small-scale anomaly classification tasks as a case, we evaluate GI-LS, and the results clearly demonstrate its superior performance. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Based Efficiently Privacy-Preserving Federated Learning Framework Using Shamir Secret Sharing

引用

IEEE Transactions on Consumer Electronics 2024年 1-1页

作者： Fu, Xiya Xiong, Ling Li, Fagen Yang, Xingchun Xiong, Naixue School of Computer and Software Engineering Xihua University Chengdu China University of Electronic Science and Technology of China Chengdu China Intelligent Policing Key Laboratory of Sichuan Province Chengdu China College of Intelligence and Computing Tianjin University Tianjin China

Federated learning, an emerging distributed learning paradigm, offers significant advantages and holds promise for addressing trust issues, breaking down data silos, and enabling active data sharing in the realm of consumer electronics. However, traditional federated learning faces critical security and privacy challenges in this domain, including single-point attacks, inference attacks, and poisoning attacks. To address these pressing security concerns, this paper proposes a decentralized federated learning framework focusing on security, reliability, and efficiency, tailored to meet the sustainability demands of consumer electronics. The proposed framework is founded upon masking and secret sharing techniques, establishing an emphatic privacy-preserving federated learning framework that ensures the security of gradient data and robustness against participant dropouts. Additionally, we actively motivate high-quality participants to collaborate by incorporating an incentive mechanism. Building upon the enhancement of existing federated learning approaches reliant on masking techniques, the method outlined in this paper significantly reduces communication overhead while preserving accuracy. Empirical research results comprehensively substantiate the superiority of this approach. Furthermore, compared to prevalent blockchain-based federated learning methods, our approach makes noteworthy strides in accuracy and efficiency. IEEE

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

SFR: Semantic-Aware Feature Rendering of Point Cloud

SFR: Semantic-Aware Feature Rendering of Point Cloud

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yaohua Zha Rongsheng Li Tao Dai Jianyu Xiong Xin Wang Shu-Tao Xia Tsinghua Shenzhen International Graduate School Tsinghua University College of Computer Science and Software Engineering Shenzhen University Peng Cheng Laboratory Research Center of Artificial Intelligence

Multi-view projection methods have demonstrated their ability to reach state-of-the-art performance in point cloud downstream tasks(e.g., classification and retrieval). These methods first require rendering the point cloud into 2D multi-view images. However, conventional methods only project the geometry of the point cloud, and such projections inevitably suffer from a loss of point cloud semantic information due to dimensionality reduction. We propose a semantic-aware and task-oriented differentiable feature rendering (SFR), which reduces the information loss during projection by generating rendered images with more point cloud semantic information for downstream tasks. Our SFR method can be applied as a plug-and-play module added to any multi-view-based backbone network for end-to-end training. Extensive experiments on benchmark datasets show that our SFR method reaches state-of-the-art performance and brings general improvements to point cloud classification and retrieval tasks.

关键词： Point cloud compression Training Geometry Three-dimensional displays Shape Semantics Signal processing

来源：评论

学校读者我要写书评

暂无评论

A LSTM and GRU-Based Hybrid Model in the Cryptocurrency Price Prediction 5th

A LSTM and GRU-Based Hybrid Model in the Cryptocurrency P...

引用

5th International Conference on Blockchain and Trustworthy Systems, BlockSys 2023

作者： Liu, Yue Xiao, Guijiao Chen, Weili Zheng, Zibin School of Computer Science and Engineering Sun Yat-sen University Guangzhou510000 China South China Business College Guangdong University of Foreign Studies Guangzhou510000 China School of Information Science and Technology Guangdong University of Foreign Studies Guangzhou510000 China School of Software Engineering Sun Yat-sen University Guangzhou510000 China

ISBN: (纸本)9789819981038

Cryptocurrency is a new type of digital currency that utilizes blockchain technology and cryptography to achieve transparency, decentralization, and immutability. Bitcoin became the world’s first decentralized cryptocurrency in 2009. With increasing attention given to cryptocurrency, predicting its price has become a popular research topic. Many machine learning and deep learning algorithms, such as Gated Recurrent Unit (GRU), Neural Network (NN), and Long Short-Term Memory (LSTM), have been studied for cryptocurrency price prediction. In this paper, we propose a hybrid cryptocurrency price prediction model based on LSTM and GRU. The model achieves better results than the LSTM model in cryptocurrency price prediction and performs the best among existing hybrid models based on LSTM and GRU. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Long short-term memory

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：