检索结果-内蒙古大学图书馆

IEEE International Conference on data Mining (ICDM)

作者： Yangcheng Gao Zhao Zhang Richang Hong Haijun Zhang Jicong Fan Shuicheng Yan School of Computer Science and Information Engineering Hefei University of Technology Hefei China Key Laboratory of Knowledge Engineering with Big Data (Ministry of Education) & Intelligent Interconnected Systems Laboratory of Anhui Province Hefei University of Technology Hefei China Department of Computer Science Harbin Institute of Technology (Shenzhen) Xili University Town Shenzhen China School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China National University of Singapore Singapore

To obtain lower inference latency and less memory footprint of deep neural networks, model quantization has been widely employed in deep model deployment, by converting the floating points to low-precision integers. However, previous methods (such as quantization aware training and post training quantization) require original data for the fine-tuning or calibration of quantized model, which makes them inapplicable to the cases that original data are not accessed due to privacy or security. This gives birth to the data-free quantization method with synthetic data generation. While current data-free quantization methods still suffer from severe performance degradation when quantizing a model into lower bit, caused by the low inter-class separability of semantic features. To this end, we propose a new and effective data-free quantization method termed ClusterQ, which utilizes the feature distribution alignment for synthetic data generation. To obtain high inter-class separability of semantic features, we cluster and align the feature distribution statistics to imitate the distribution of real data, so that the performance degradation is alleviated. Moreover, we incorporate the diversity enhancement to solve class-wise mode collapse. We also employ the exponential moving average to update the centroid of each cluster for further feature distribution improvement. Extensive experiments based on different deep models (e.g., ResNet-18 and MobileNet-V2) over the ImageNet dataset demonstrate that our proposed ClusterQ model obtains state-of-the-art performance.

关键词： Degradation Training Quantization (signal) Semantics Neural networks Memory management data models

来源：评论

学校读者我要写书评

暂无评论

FineFormer: Fine-Grained Adaptive Object Transformer for Image Captioning

FineFormer: Fine-Grained Adaptive Object Transformer for Ima...

引用

IEEE International Conference on data Mining (ICDM)

作者： Bo Wang Zhao Zhang Jicong Fan Mingbo Zhao Choujun Zhan Mingliang Xu School of Computer Science and Information Engineering Hefei University of Technology Hefei China Key Laboratory of Knowledge Engineering with Big Data (Ministry of Education) & Intelligent Interconnected Systems Laboratory of Anhui Province Hefei University of Technology Hefei China School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China School of Information Science and Technology Donghua University Shanghai China School of Electrical and Computer Engineering Nanfang College Guangzhou Guangzhou China School of Information Engineering Zhengzhou University Zhengzhou China

Image captioning is still a challenging task aiming at describing the contents of image by words. Current image caption methods usually assume the object relation to be important if the semantic and spatial geometric relationships between objects are close and large, but the relations meeting this assumption are not necessarily important to describe the contents of image in a fine-grained way. That is, the importance of fine-grained object relations is not properly taken into account. Besides, current Transformer based image caption models also fail to consider the importance of fine-grained objects, since they generate all the words of a sentence at one time, which cannot Figure out which objects are more important and vice versa. In this paper, we propose a novel Fine-grained Adaptive Object Transformer (FineFormer) network, which can jointly discover the importance of fine-grained objects and object relations for image captioning. Specifically, a new concept of adaptive soft-foreground attention is proposed to highlight the fine-grained objects dominating the descriptive contents. To characterize and calculate the important relations between fine-grained objects, we also propose an adaptive object relation attention to refine the object relation from the generation process of relation. As such, FineFormer can describe the contents of image more accurately, by reducing the interference of unimportant objects in the background. Extensive experiments on the highly-competitive MS-COCO dataset demonstrated the superiority of our FineFormer.

关键词： Adaptation models Adaptive systems Current transformers Semantics Interference data mining Task analysis

来源：评论

学校读者我要写书评

暂无评论

ReconBoost: boosting can achieve modality reconcilement 24

ReconBoost: boosting can achieve modality reconcilement

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Cong Hua Qianqian Xu Shilong Bao Zhiyong Yang Qingming Huang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China and School of Cyber Security University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China and Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores a novel multi-modal alternating learning paradigm pursuing a reconciliation between the exploitation of uni-modal features and the exploration of cross-modal interactions. This is motivated by the fact that current paradigms of multi-modal learning tend to explore multi-modal features simultaneously. The resulting gradient prohibits further exploitation of the features in the weak modality, leading to modality competition, where the dominant modality overpowers the learning process. To address this issue, we study the modality-alternating learning paradigm to achieve reconcilement. Specifically, we propose a new method called ReconBoost to update a fixed modality each time. Herein, the learning objective is dynamically adjusted with a reconcilement regularization against competition with the historical models. By choosing a KL-based reconcilement, we show that the proposed method resembles Friedman's Gradient-Boosting (GB) algorithm, where the updated learner can correct errors made by others and help enhance the overall performance. The major difference with the classic GB is that we only preserve the newest model for each modality to avoid overfitting caused by ensembling strong learners. Furthermore, we propose a memory consolidation scheme and a global rectification scheme to make this strategy more effective. Experiments over six multi-modal benchmarks speak to the efficacy of the method. We release the code at https://***/huacong/ReconBoost.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Asymmetric Cross-Scale Alignment for Text-Based Person Search

arXiv

引用

arXiv 2022年

作者： Ji, Zhong Hu, Junhua Liu, Deyin Wu, Lin Yuanbo Zhao, Ye The School of Electrical and Information Engineering Tianjin University Tianjin300072 China Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Artificial Intelligence Anhui University Hefei230039 China Department of Computer Science Swansea University SA1 8EN United Kingdom The Key Laboratory of Knowledge Engineering with Big Data School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China

Text-based person search (TBPS) is of significant importance in intelligent surveillance, which aims to retrieve pedestrian images with high semantic relevance to a given text description. This retrieval task is characterized with both modal heterogeneity and fine-grained matching. To implement this task, one needs to extract multi-scale features from both image and text domains, and then perform the cross-modal alignment. However, most existing approaches only consider the alignment confined at their individual scales, e.g., an image-sentence or a region-phrase scale. Such a strategy adopts the presumable alignment in feature extraction, while overlooking the cross-scale alignment, e.g., image-phrase. In this paper, we present a transformer-based model to extract multi-scale representations, and perform Asymmetric Cross-Scale Alignment (ACSA) to precisely align the two modalities. Specifically, ACSA consists of a global-level alignment module and an asymmetric cross-attention module, where the former aligns an image and texts on a global scale, and the latter applies the cross-attention mechanism to dynamically align the cross-modal entities in region/image-phrase scales. Extensive experiments on two benchmark datasets CUHK-PEDES and RSTPReid demonstrate the effectiveness of our approach. Codes are available at https://***/mulhjh/ACSA. Copyright © 2022, The Authors. All rights reserved.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Exploring linguistic features and user engagement in Chinese online mental health counseling

引用

Heliyon 2024年第19期10卷 e38042页

作者： Zhang, Liyuan Liu, Dexi Li, Jing Wan, Changxuan Liu, Xiping School of Computer and Artificial Intelligence Jiangxi University of Finance and Economics JiangXi Nanchang 330013 China School of Mathematics and Computer YuZhang Normal College Jiangxi Nanchang 330013 China Jiangxi Key Laboratory of Data and Knowledge Engineering Jiangxi University of Finance and Economics Jiangxi Nanchang 330013 China School of Electronic Management Science Fujian Jiangxia University Fujian Fuzhou 350108 China

With the popularity of online mental health platforms, more individuals are seeking help and receiving social support by openly discussing their problems. Therefore, it's crucial to gain a deeper understanding of which problem disclosures and social support on these platforms can attract more user attention and engagement. Previous research has primarily focused on social media forums. Our work concentrates on the professional mental health platform, intending to understand the linguistic features present in posts that promote user engagement and interaction. We employ text mining and deep learning techniques to analyze posts consisting of 22,250 questions from help-seekers and 78,328 answers providing social support extracted from the Chinese online mental health counseling platform. Initially, we analyze the high-frequency words and topics of the questions and answers to gain insights into the primary focal points and the range of topics covered in these posts. The results indicate that work-related issues are the most concerning and troublesome for help-seekers, and the topics that users follow are approximately 8 types, including growth, family, in-love, marriage, emotions, human-relations, behavioral-therapy and career. Subsequently, we analyze the language usage in question-and-answer posts with different engagement from three aspects: vocabulary categories, linguistic style matching, and language modeling, aiming to identify which linguistic features can attract more user attention and engagement. The results reveal that high-engagement answer posts exhibit a higher degree of linguistic style matching with the corresponding questions, and the use of vocabulary categories also influences the attention and engagement of the posts. By exploring the linguistic features and patterns displayed in posts with different levels of engagement on the professional online mental health platform, this study offers deep insights into user behavior and the factors that impact

关键词： Language modeling Linguistic style matching Online mental health counseling The topics of questions and answers Vocabulary categories

来源：评论

学校读者我要写书评

暂无评论

Multi-fuzzy-constrained graph pattern matching with big graph data

引用

Intelligent data Analysis 2020年第4期24卷 941-958页

作者： Liu, Guliu Li, Lei Wu, Xindong Key Laboratory of Knowledge Engineering with Big Data (Hefei University of Technology) Ministry of Education Hefei Anhui 230009 China School of Computer Science and Information Engineering Hefei University of Technology Hefei Anhui China Mininglamp Academy of Sciences Mininglamp Technologies Beijing China

Graph pattern matching has been widespread used for protein structure analysis, expert finding and social group selection, ect. Recently, the study of graph pattern matching using the abundant attribute information of vertices and edges as constraint conditions has attracted the attention of scholars, and multi-constrained simulation has been proposed to address the problem in contextual social networks. Actually, multi-constrained graph pattern matching is an NP-complete problem and the fuzziness of constraint variables may exist in many applications. In this paper, we introduce a multi-fuzzy-constrained graph pattern matching problem in big graph data, and propose an efficient first-k algorithm Fuzzy-ETOF-K for solving it. Specifically, exploration-based method based on edge topology is adopted to improve the efficiency of edge connection, and breadth-first bounded search is used for edge matching instead of shortest path query between two nodes to improve the efficiency of edge matching. The results of our experiments conducted on three datasets of real social networks illustrate that our proposed algorithm Fuzzy-ETOF-K significantly outperforms existing approaches in efficiency and the introduction of fuzzy constraints makes our proposed algorithm more efficient and effective. © 2020 - IOS Press and the authors. All rights reserved.

关键词： big graph data Graph pattern matching multi-fuzzy-constrained

来源：评论

学校读者我要写书评

暂无评论

UAV-enabled Collaborative Beamforming via Multi-Agent Deep Reinforcement Learning

arXiv

引用

arXiv 2024年

作者： Liu, Saichao Sun, Geng Li, Jiahui Liang, Shuang Wu, Qingqing Wang, Pengfei Niyato, Dusit The College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Changchun130012 China The College of Computing and Data Science Nanyang Technological University 639798 Singapore The School of Information Science and Technology Northeast Normal University Changchun130117 China The Department of Electronic Engineering Shanghai Jiao Tong University Shanghai200240 China The School of Computer Science and Technology Dalian University of Technology Dalian116024 China

In this paper, we investigate an unmanned aerial vehicle (UAV)-assistant air-to-ground communication system, where multiple UAVs form a UAV-enabled virtual antenna array (UVAA) to communicate with remote base stations by utilizing collaborative beamforming. To improve the work efficiency of the UVAA, we formulate a UAV-enabled collaborative beamforming multi-objective optimization problem (UCBMOP) to simultaneously maximize the transmission rate of the UVAA and minimize the energy consumption of all UAVs by optimizing the positions and excitation current weights of all UAVs. This problem is challenging because these two optimization objectives conflict with each other, and they are non-concave to the optimization variables. Moreover, the system is dynamic, and the cooperation among UAVs is complex, making traditional methods take much time to compute the optimization solution for a single task. In addition, as the task changes, the previously obtained solution will become obsolete and invalid. To handle these issues, we leverage the multi-agent deep reinforcement learning (MADRL) to address the UCBMOP. Specifically, we use the heterogeneous-agent trust region policy optimization (HATRPO) as the basic framework, and then propose an improved HATRPO algorithm, namely HATRPO-UCB, where three techniques are introduced to enhance the performance. Simulation results demonstrate that the proposed algorithm can learn a better strategy compared with other methods. Moreover, extensive experiments also demonstrate the effectiveness of the proposed techniques. Copyright © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

UAV Virtual Antenna Array Deployment for Uplink Interference Mitigation in data Collection Networks

arXiv

引用

arXiv 2024年

作者： Li, Hongjuan Kang, Hui Sun, Geng Li, Jiahui Wang, Jiacheng Wang, Xue Niyato, Dusit Leung, Victor C.M. College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China College of Computing and Data Science Nanyang Technological University Singapore639798 Singapore College of Computing and Data Science Nanyang Technological University Singapore Department of Communication Engineering Jilin University Changchun130012 China Artificial Intelligence Research Institute Shenzhen MSU-BIT University Shenzhen518115 China College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Electrical and Computer Engineering The University of British Columbia VancouverV6T 1Z4 Canada

Unmanned aerial vehicles (UAVs) have gained considerable attention as a platform for establishing aerial wireless networks and communications. However, the line-of-sight dominance in air-to-ground communications often leads to significant interference with terrestrial networks, reducing communication efficiency among terrestrial terminals. This paper explores a novel uplink interference mitigation approach based on the collaborative beamforming (CB) method in multi-UAV network systems. Specifically, the UAV swarm forms a UAV-enabled virtual antenna array (VAA) to achieve the transmissions of gathered data to multiple base stations (BSs) for data backup and distributed processing. However, there is a trade-off between the effectiveness of CB-based interference mitigation and the energy conservation of UAVs. Thus, by jointly optimizing the excitation current weights and hover position of UAVs as well as the sequence of data transmission to various BSs, we formulate an uplink interference mitigation multi-objective optimization problem (MOOP) to decrease interference affection, enhance transmission efficiency, and improve energy efficiency, simultaneously. In response to the computational demands of the formulated problem, we introduce an evolutionary computation method, namely chaotic non-dominated sorting genetic algorithm II (CNSGA-II) with multiple improved operators. The proposed CNSGA-II efficiently addresses the formulated MOOP, outperforming several other comparative algorithms, as evidenced by the outcomes of the simulations. Moreover, the proposed CB-based uplink interference mitigation approach can significantly reduce the interference caused by UAVs to non-receiving BSs. Copyright © 2024, The Authors. All rights reserved.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Graph-Based Terminal Ranking for Sparsification of Interference Coordination Parameters in Ultra-Dense Networks 10th

Graph-Based Terminal Ranking for Sparsification of Interfere...

引用

10th EAI International Conference on Mobile Networks and Management, MONAMI 2020

作者： Gou, Junxiang Wang, Lusheng Lin, Hai Peng, Min Ministry of Education School of Computer Science and Information Engineering Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Hefei230601 China Ministry of Education School of Cyber Science and Engineering Key Laboratory of Aerospace Information Security and Trusted Computing Wuhan University Wuhan430072 China Anhui Province Key Laboratory of Industry Safety and Emergency Technology Hefei230601 China

ISBN: (纸本)9783030640019

In the future mobile communication system, inter-cell interference becomes a serious problem due to the intensive deployment of cells and terminals. Traditional interference coordination schemes take long time for optimization in ultra-dense networks. Meanwhile, due to the increase of factors affecting communication and in order to better meet the communication needs of each terminal, an interference coordination scheme needs to fully consider multiple characteristic parameters of the terminal, which will further increase the scheme’s computational time. Therefore, we should compress all the data through sparsification of parameters before optimization. There are many terminal parameters, and the essence of sparsification of parameters is to rank terminals. In this paper, a graph-based terminal ranking scheme is designed. First, each terminal can be represented by its multiple parameters. Then, all terminals are used as the vertexes in the graph to form a complete weighted graph, and edge weights represent the degree of dissimilarity between terminals. A ranking of terminals is obtained by finding a minimum Hamiltonian path in the graph. Finally, the ranking of all parameter sequences is obtained according to terminals ranking, which makes the sparsity of all parameter sequences better. Simulation results show that the proposed scheme can accomplish sparsification of parameter sequences effectively, especially when the number of sequences increases. In addition, compared with the optimal coordination of traditional scheme, this scheme improves the fairness of the system while ensuring high system capacity, and dramatically reduces the computational time of interference coordination. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications engineering 2020.

关键词： Approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Multi-scale Graph Convolutional Neural Network for Object Recognition from Point Cloud data 2nd

Multi-scale Graph Convolutional Neural Network for Object Re...

引用

2nd International Conference on Urban Intelligence and Applications, ICUIA 2020

作者： Lu, Qiang Chen, Chao Teng, Jinfeng Zhang, Chunyuan Huang, Yi Xuan, Shanli Key Laboratory of Knowledge Engineering with Big Data Ministry of Education Hefei University of Technology Hefei230009 China School of Computer and Information Hefei University of Technology Hefei230009 China Anhui Province Key Laboratory of Industry Safety and Emergency Technology Hefei230009 China

ISBN: (纸本)9789813346000

How to make robots understand the point cloud data which is collected from the 3D sensor and complete the recognition has become a hot research direction in recent years. In this paper, we propose a new approach to improve the critical robotic capability, semantic understanding of the environment (i.e., 3D object recognition). The convolutional neural network (CNN) method has a very good recognition result in the 2D image domain, but it has certain difficulty in applying irregular and unordered 3D point clouds data. The network for point cloud data generally uses the convolution to realize the extraction of point cloud features by finding the neighborhood features on the point set. Due to the different neighborhood scales caused by the irregularity of 3D point cloud data, we propose a CNN structure that combines multi-scale features. By finding multiple neighborhoods of the point set and establishing local graph extraction features, the stable expression of the local neighborhood is obtained. At the same time, the key point calibration method is added, so that the network can dynamically focus on key point features to improve the recognition result. In a series of analytical experiments, we demonstrate competing results that demonstrate the effectiveness of the network structure. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Convolutional neural networks

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：