检索结果-内蒙古大学图书馆

2023 International Conference on Algorithms, High Performance Computing, and Artificial Intelligence, AHPCAI 2023

作者： Zhang, Chenyu Ye, Qihong Chen, Hongming Wang, Xiaoshuang School of Information Engineering Zhejiang Ocean University Zhoushan China Key Laboratory of Oceanographic Big Data Mining Application of Zhejiang Province China

ISBN: (纸本)9781510671881

In recent years, significant progress has been made in image dehazing, but most dehazing convolutional neural networks only learn from hazy images to the corresponding feature maps of clean images, ignoring the details of the images. In this paper, a new flatness loss function is proposed for single-image dehazing, thereby improving the overall effect of dehazing results. This loss function allows the model to focus on the texture features of hazy images and clean images, and supervises edge information by reducing the flatness difference between clean pixels and blurred pixels. The experiments on the benchmark dataset using the flatness loss function on the single-image dehazing model show that this method can effectively improve the quantitative performance of the model. © 2023 SPIE. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Solving Multiobjective Combinatorial Optimization via Learning to Improve Method

引用

IEEE Transactions on Emerging Topics in Computational Intelligence 2025年第3期9卷 2122-2136页

作者： Ye, Te Zhang, Zizhen Zhang, Qingfu Chen, Jinbiao Wang, Jiahai Sun Yat-sen University School of Computer Science and Engineering Guangzhou510275 China Sun Yat-sen University Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510275 China City University of Hong Kong Department of Computer Science Kowloon Hong Kong

Recently, neural combinatorial optimization (NCO) methods have been prevailing for solving multiobjective combinatorial optimization problems (MOCOPs). Most NCO methods are based on the "Learning to Construct"(L2C) paradigm, where the trained model(s) can directly generate a set of approximate Pareto optimal solutions. However, these methods still suffer from insufficient proximity and poor diversity towards the true Pareto front. In this paper, following the "Learning to Improve"(L2I) paradigm, we propose weight-related policy network (WRPN), a learning-based improvement method for solving MOCOPs. WRPN is incorporated into multiobjective evolutionary algorithm (MOEA) frameworks to effectively guide the search direction. A shared baseline for proximal policy optimization is presented to reduce variance in model training. A quality enhancement mechanism is designed to further refine the Pareto set during model inference. Computational experiments conducted on two classic MOCOPs, i.e., multiobjective traveling salesman problem and multiobjective vehicle routing problem, indicate that our method achieves remarkable results. Notably, our WRPN module can be easily integrated into various MOEA frameworks such as NSGA-II, MOEA/D and MOGLS, providing versatility and applicability across different problem domains. © 2025 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Multi-scale cyclical similarity prototype refinement for few-shot breast ultrasound image segmentation 17

Multi-scale cyclical similarity prototype refinement for few...

引用

17th IEEE International Conference on Signal Processing, ICSP 2024

作者： Ou, Yingfeng Yang, Xing Zhang, Jian Jian, Caiqing Wang, Lihui College of Computer Science and Technology Guizhou University Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province State Key Laboratory of Public Big Data Engineering Research Center of Text Computing Ministry of Education Guiyang China

ISBN: (纸本)9798350387384

Few-shot learning based methods can address the reliance on large-scale labeled samples in current breast tumor segmentation. However, previous methods typically rely on a few support samples to extract abstract, coarse-grained semantic prototypes, without considering the interaction between support and query data. In this study, we aim to provide each query image with relevant and information-rich support information. By integrating the difficult point selection module and the multi-scale cyclical prototype refinement module, we progressively refine coarse-grained support prototypes into more informative, fine-grained class prototypes, thereby better matching the highly variable breast ultrasound query data. Experiments on two datasets demonstrate that our proposed multi-scale cyclical similarity prototype refinement network (CSPRNet) achieves remarkable improvements compared to previous state-of-the-art methods. Specifically, using ResNet50 as the backbone, we achieve 57.93%/57.92% and 57.58%/59.24% mIoU for 1-shot and 5-shot segmentation, respectively, on public and private datasets, outperforming the suboptimal method by 8.18%/2.55% and 2.77%/4.48%. © 2024 IEEE.

关键词： Ultrasonic imaging

来源：评论

学校读者我要写书评

暂无评论

A Positive Influence Maximization Algorithm in Signed Social Networks

引用

Computers, Materials & Continua 2023年第8期76卷 1977-1994页

作者： Wenlong Zhu Yang Huang Shuangshuang Yang Yu Miao Chongyuan Peng College of Computer and Control Engineering Qiqihar UniversityQiqihar161006China Heilongjiang Key Laboratory of Big Data Network Security Detection and Analysis Qiqihar UniversityQiqihar161006China College of Teacher Education Qiqihar UniversityQiqihar161006China

The influence maximization(IM)problem aims to find a set of seed nodes that maximizes the spread of their influence in a social *** positive influence maximization(PIM)problem is an extension of the IM problem,which consider the polar relation of nodes in signed social networks so that the positive influence of seeds can be the most widely *** solve the PIM problem,this paper proposes the polar and decay related independent cascade(IC-PD)model to simulate the influence propagation of nodes and the decay of information during the influence propagation in signed social *** overcome the low efficiency of the greedy based algorithm,this paper defines the polar reverse reachable(PRR)set and devises a signed reverse influence sampling(SRIS)*** algorithm utilizes the ICPD model as well as the PRR set to select *** are two phases in *** is the sampling phase,which utilizes the IC-PD model to generate the PRR set and a binary search algorithm to calculate the number of needed PRR *** other is the node selection phase,which uses a greedy coverage algorithm to select optimal ***,Experiments on three real-world polar social network datasets demonstrate that SRIS outperforms the baseline algorithms in *** on the Slashdot dataset,SRIS achieves 24.7% higher performance than the best-performing compared algorithm under the weighted cascade model when the seed set size is 25.

关键词： Positive influencemaximization polar relation information decay reverse influence sampling

来源：评论

学校读者我要写书评

暂无评论

Predicting ICU Pressure Injuries with Historical data: A Multivariate Time Series Approach

Predicting ICU Pressure Injuries with Historical Data: A Mul...

引用

IEEE International Conference on Big knowledge (ICBK)

作者： Lintai Cui Liuqi Jin Key Laboratory of Knowledge Engineering with Big Data of the Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology Hefei China

Hospital-acquired pressure injury (HAPI) is a se-rious healthcare problem for intensive care unit (ICU) patients, which significantly affects their quality of life and prognosis, and increases hospitalization time and medical expenses. The current prediction models cannot accurately predict the pressure injury (PI) in ICU wards, as these models usually only predict based on the patient's current physical condition and electronic health record data, resulting in poor recall and precision, which affects the prediction performance. To solve this problem, we applied the multivariate time series data of I CU patients from admission to discharge and established a PI prediction model for ICU patients by utilizing the bidirectional long short-term memory neural network (Bi-LSTM) model. Experiments on the MIMIC-III (Medical Information Mart for Intensive Care) dataset show that the Bi-LSTM model has an F1 score of 0.24 and an AUC (Area Under Curve) value of 0.81, which are better than other models. This validates the effectiveness of the Bi-LSTM on the multivariate time series data to predict the occurrence of PI in ICU wards and assist nursing staff to allocate nursing resources more effectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

UARC:Unsupervised Anomalous Traffic Detection with Improved U-Shaped Autoencoder and RetNet Based Multi-clustering 26th

UARC:Unsupervised Anomalous Traffic Detection with Improved ...

引用

26th International Conference on Information and Communications Security, ICICS 2024

作者： Xie, Yunyang Chen, Kai Li, Shenghui Li, Bingqian Zhang, Ning Hubei Key Laboratory of Distributed System Security Hubei Engin eering Research Center on Big Data Security School of Cyber Science and Engineering Huazhong University Technology Wuhan430074 China

ISBN: (纸本)9789819787975

With the ongoing advancement of deep learning, modern network intrusion detection systems increasingly favor utilizing deep learning networks to improve their ability to learn traffic characteristics. To address the challenge of obtaining a substantial amount of labeled training data, many intrusion detection systems now focus on unsupervised anomaly detection methods. Despite this shift, researchers still face the daunting task of distinguishing a significant volume of anomalous traffic and dealing with data imbalance. To address these real-world challenges, we introduce UARC, a system capable of achieving unsupervised anomaly traffic detection through multi-clustering. UARC utilizes an enhanced U-shaped autoencoder and a feature fusion method incorporating Masked Retnet to effectively extract spatiotemporal features from network traffic. It combines these techniques with the HDBSCAN algorithm for multi-clustering of traffic, providing a form of reverse guidance for network learning. Experimental results on multiple datasets demonstrate that UARC can cluster various types of traffic with an impressive accuracy rate of up to 97.96%, while achieving a 99.70% AUC value for anomaly detection. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Attribution rollout: a new way to interpret visual transformer

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第1期14卷 163-173页

作者： Xu, Li Yan, Xin Ding, Weiyue Liu, Zechao College of Computer Science and Technology Harbin Engineering University Nantong Street Heilongjiang Harbin150001 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Qianjin Street Jilin Changchun130012 China Department of Medicine Harvard Medical School Longwood Avenue BostonMA02115 United States

Transformer-based models are dominating the field of natural language processing and are becoming increasingly popular in the field of computer vision. However, the black box characteristics of transformers seriously hamper their application in certain fields. Prior work relies on the raw attention scores or employs heuristic propagation along with the attention graph. In this work, we propose a new way to visualize model. The method computes attention scores based on attribution and then propagates these attention scores through the layers. This propagation involves attention layers and multi-head attention mechanism. Our method extracts salient dependencies in each layer to visualize prediction results. We benchmark our method on recent visual transformer networks and demonstrate its many advantages over the existing interpretability methods. Our code is available at: https://***/yxheartipp/attr-rollout. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

A Multiple-Blockchains based Service Monitoring Framework in Edge-Cloud Computing 23

A Multiple-Blockchains based Service Monitoring Framework in...

引用

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on data Science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Fu, Hang Wang, Puwei Li, Haoran Zhan, Ying Chen, Jinchuan Du, Xiaoyong Guizhou University of Finance and Economics Guiyang City550025 China School of Information Renmin University of China China Renmin University of China Key Laboratory of Data Engineering and Knowledge Engineering of Ministry of Education Beijing100872 China

ISBN: (纸本)9781665494571

In the edge-cloud computing, the applications usually are delivered as services, each of which runs independently and can cooperate to construct the complicated applications. However, it is difficult to monitor the services in the decentralized edge-cloud environment since there usually does not exist the central authorization to guarantee trustiness of the service quality data collected. To address the trust issue, the blockchain techniques have been introduced to the service monitoring. Considering the services usually are distributed on the edge nodes, if we deploy only a blockchain on the edge nodes near to the services, it may have to support both the service registration and monitoring because of the loose-coupling principle. That would influence the performance, even would crash the edge nodes. In this paper, we propose a multiple-blockchains based framework, including a registration blockchain that focuses on the service registration and discovery, and a set of monitoring blockchains that focus on the service invocation and monitoring. The registration blockchain is deployed on the cloud nodes, and the monitoring blockchains are deployed on the edge nodes near to the services for collecting efficiently and precisely the service quality data. Based on the cross-chain technology, we design a cross-chain service discovery between the monitoring blockchain and the registration blockchain for finding the services the requesters need. Finally, we implement the registration and monitoring blockchains based on the blockchain platform Cosmos and conduct a set of experiments to show the effectiveness of the proposed framework. © 2021 IEEE.

关键词： Authorization Cloud computing Smart cities Computer crashes Blockchains Monitoring

来源：评论

学校读者我要写书评

暂无评论

Resolution-Adaptive Representation Learning via Eliminating Abnormal Features for Cross-Resolution Person Re-Identification 14

Resolution-Adaptive Representation Learning via Eliminating ...

引用

14th International Conference on Information Technology in Medicine and Education, ITME 2024

作者： Lu, Xing Lai, Fengshan Cao, Zhixiang Xia, Daoxun Guizhou Normal University School of Big Data and Computer Science Guiyang China Guizhou Normal University Key Laboratory of Information and Computing Science Guizhou Province Guiyang China Guizhou Normal University Engineering Laboratory for Applied Technology of Big Data in Education Guiyang China

ISBN: (纸本)9798350356670

Cross-resolution person re-identification(CR-ReID) seeks to overcome the challenge of retrieving and matching specific person images across cameras with varying resolutions. Numerous existing studies utilize established CNNs and ViTs models to resize captured low-resolution (LR) images and align them with high-resolution (HR) image features. However, these methods overlook the potential feature connections between LR and HR images. Consequently, CNNs or ViTs tend to highlight outliers rather than the intended characteristics in LR image attention maps. In this work, we propose the abnormal feature elimination and reconfiguration transformer (ART) network, a novel network architecture, for robust cross-resolution person re-identification tasks. Our proposed method uses a resolution feature discriminator to learn resolution invariant features and outputs feature matrices of images with different resolutions. We utilize output feature matrices to model LR and HR image interactions through cross-resolution landmark agent attention. We aim to mitigate abnormal image features and prioritize the attention on the target person by learning representations from input images of various resolutions. Extensive evaluations on three real-world datasets confirm the excellent results of our approach. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

FlexPDA:A Flexible Programming Framework for Deep Learning Accelerators

引用

Journal of Computer Science & Technology 2022年第5期37卷 1200-1220页

作者： Lei Liu Xiu Ma Hua-xiao Liu Guang-li Li Lei Liu College of Computer Science and Technology Jilin UniversityChangchunChina Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin UniversityChangchunChina State Key Laboratory of Computer Architecture Institute of Computing TechnologyChinese Academy of SciencesBeijingChina University of Chinese Academy of Sciences BeijingChina

There are a wide variety of intelligence accelerators with promising performance and energy efficiency,deployed in a broad range of applications such as computer vision and speech ***,programming productivity hinders the deployment of deep learning *** low-level library invoked in the high-level deep learning framework which supports the end-to-end execution with a given model,is designed to reduce the programming burden on the intelligence ***,it is inflexible for developers to build a network model for every deep learning application,which probably brings unnecessary repetitive *** this paper,a flexible and efficient programming framework for deep learning accelerators,FlexPDA,is proposed,which provides more optimization opportunities than the low-level library and realizes quick transplantation of applications to intelligence accelerators for fast *** evaluate FlexPDA by using 10 representative operators selected from deep learning algorithms and an end-to-end *** experimental results validate the effectiveness of FlexPDA,which achieves an end-to-end performance improvement of 1.620x over the low-level library.

关键词： deep learning accelerator programming framework domain-specific language

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：