检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Zhang, Chuan-Fei Liu, Yan Zhang, Tian-Hao Chen, Song-Lu Chen, Feng Yin, Xu-Cheng School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing100083 China Key Laboratory of Knowledge Automation for Industrial Processes Ministry of Education Beijing100083 China USTB-EEasyTech Joint Lab of Artificial Intelligence University of Science and Technology Beijing Beijing100083 China EEasy Technology Company Ltd. Zhuhai519000 China

Non-autoregressive (NAR) transformer models have been studied intensively in automatic speech recognition (ASR), and a substantial part of NAR transformer models is to use the casual mask to limit token dependencies. However, the casual mask is designed for the left-to-right decoding process of the non-parallel autoregressive (AR) transformer, which is inappropriate for the parallel NAR transformer since it ignores the right-to-left contexts. Some models are proposed to utilize right-to-left contexts with an extra decoder, but these methods increase the model complexity. To tackle the above problems, we propose a new non-autoregressive transformer with a unified bidirectional decoder (NAT-UBD), which can simultaneously utilize left-to-right and right-to-left contexts. However, direct use of bidirectional contexts will cause information leakage, which means the decoder output can be affected by the character information from the input of the same position. To avoid information leakage, we propose a novel attention mask and modify vanilla queries, keys, and values matrices for NAT-UBD. Experimental results verify that NAT-UBD can achieve character error rates (CERs) of 5.0%/5.5% on the Aishell1 dev/test sets, outperforming all previous NAR transformer models. Moreover, NAT-UBD can run 49.8× faster than the AR transformer baseline when decoding in a single step. © 2021, CC BY-NC-ND.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

A novel method for road centerline extraction from high spatial resolution satellite images

引用

Journal of Information Hiding and Multimedia Signal Processing 2016年第5期7卷 1069-1080页

作者： Xu, Rui Key Lab of Spatial Data Mining & Information Sharing of Ministry of Education Spatial Information Research Center of Fujian Province Fuzhou University No.523 Gongye Road Fuzhou350002 China College of Information Science and Engineering Fujian University of Technology No.3 Xueyuan Road University Town Fuzhou350108 China Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology No.3 Xueyuan Road University Town Fuzhou350108 China

An integrated method for road centerline extraction has been proposed in this paper which consists of four main steps. First, the Graph Cut algorithm using proposed similarity function is employed to segment the image, and the segmented image is thresholded to initial binary skeleton. Second, the road skeleton is refined by using a series of morphology operations and shape features. Third, the curvilinear feature of road skeleton is detected by a set of multiple filtering detectors. Finally,a regression method is performed to extract smooth and accurate road centerlines. Validation tests show that the proposed method is able to achieve a comparatively good performance in road centerline extraction from high spatial resolution satellite images. © 2016, Ubiquitous International. All Rights Reserved.

关键词： Regression analysis

来源：评论

学校读者我要写书评

暂无评论

A hierarchical object-based method for suburban road extraction from remote sensing image

引用

ICIC Express Letters, Part B: Applications 2016年第12期7卷 2671-2676页

作者： Xu, Rui College of Information Science and Engineering Fujian University of Technology No. 3 Xueyuan Road University Town Fuzhou350118 China Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology No. 3 Xueyuan Road University Town Fuzhou350118 China Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education Spatial Information Research Center of Fujian Province Fuzhou University No. 523 Gongye Road Fuzhou350002 China

Road extraction from remote sensing image is a very popular topic. There are many methods dealing with this problem, but few of them focus on suburban roads. This paper presents a hierarchical object-based method for suburban road extraction. First, the image is preprocessed by the Normalized Difference Vegetation Index (NDVI) and the Hue-Saturation-Intensity (HSI) Index. Vegetation regions, shadows and water areas are then removed. Second, texture features are applied to classifying the preprocessed image into two categories: road and non-road. Homogeneous property is combined to remove false classification so as to generate initial road skeleton. Third, the road skeleton is refined by using a series of morphology operations and shape features. Fourth, the curvilinear feature of road skeleton is detected by a set of multiple filtering detectors. Finally, a regression method is performed to extract smooth and accurate road centerlines. The experimental results indicate that the proposed method is suitable for suburban road extraction. © 2016 ISSN 2185-2766.

关键词： Vegetation

来源：评论

学校读者我要写书评

暂无评论

ZSTAD: Zero-shot temporal activity detection

arXiv

引用

arXiv 2020年

作者： Zhang, Lingling Chang, Xiaojun Liu, Jun Luo, Minnan Wang, Sen Ge, Zongyuan Hauptmann, Alexander School of Computer Science and Technology Xian Jiaotong University Xian China Ministry of Education Key Lab For Intelligent Networks and Network Security Xian China Faculty of Information Technology Monash University Australia National Engineering Lab for Big Data Analytics Xian Jiaotong University Xian China School of Information Technology and Electrical Engineering University of Queensland Australia School of Computer Science Carnegie Mellon University United States

An integral part of video analysis and surveillance is temporal activity detection, which means to simultaneously recognize and localize activities in long untrimmed videos. Currently, the most effective methods of temporal activity detection are based on deep learning, and they typically perform very well with large scale annotated videos for training. However, these methods are limited in real applications due to the unavailable videos about certain activity classes and the time-consuming data annotation. To solve this challenging problem, we propose a novel task setting called zero-shot temporal activity detection (ZSTAD), where activities that have never been seen in training can still be detected. We design an end-to-end deep network based on R-C3D as the architecture for this solution. The proposed network is optimized with an innovative loss function that considers the embeddings of activity labels and their super-classes while learning the common semantics of seen and unseen activities. Experiments on both the THUMOS14 and the Charades datasets show promising performance in terms of detecting unseen activities. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Toward Enhanced Robustness in Unsupervised Graph Representation Learning: A Graph Information Bottleneck Perspective

arXiv

引用

arXiv 2022年

作者： Wang, Jihong Luo, Minnan Li, Jundong Liu, Ziqi Zhou, Jun Zheng, Qinghua The Ministry of Education Key Lab for Intelligent Networks and Network Security School of Computer Science and Technology Xi’an Jiaotong University Xi’an710049 China The Department of Electrical and Computer Engineering Department of Computer Science School of Data Science University of Virginia United States The Ant Financial Services Group Zhejiang Hangzhou310000 China

Recent studies have revealed that GNNs are vulnerable to adversarial attacks. Most existing robust graph learning methods measure model robustness based on label information, rendering them infeasible when label information is not available. A straightforward direction is to employ the widely used Infomax technique from typical Unsupervised Graph Representation Learning (UGRL) to learn robust unsupervised representations. Nonetheless, directly transplanting the Infomax technique from typical UGRL to robust UGRL may involve a biased assumption. In light of the limitation of Infomax, we propose a novel unbiased robust UGRL method called Robust Graph Information Bottleneck (RGIB), which is grounded in the Information Bottleneck (IB) principle. Our RGIB attempts to learn robust node representations against adversarial perturbations by preserving the original information in the benign graph while eliminating the adversarial information in the adversarial graph. There are mainly two challenges to optimize RGIB: 1) high complexity of adversarial attack to perturb node features and graph structure jointly in the training procedure;2) mutual information estimation upon adversarially attacked graphs. To tackle these problems, we further propose an efficient adversarial training strategy with only feature perturbations and an effective mutual information estimator with subgraph-level summary. Moreover, we theoretically establish a connection between our proposed RGIB and the robustness of downstream classifiers, revealing that RGIB can provide a lower bound on the adversarial risk of downstream classifiers. Extensive experiments over several benchmarks and downstream tasks demonstrate the effectiveness and superiority of our proposed method. Copyright © 2022, The Authors. All rights reserved.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Adaptive Structure-Constrained Robust Latent Low-Rank Coding for Image Recovery

Adaptive Structure-Constrained Robust Latent Low-Rank Coding...

引用

IEEE International Conference on data Mining (ICDM)

作者： Zhao Zhang Lei Wang Sheng Li Yang Wang Zheng Zhang Zhengjun Zha Meng Wang School of Computer Science and Technology Soochow University Suzhou China Key Laboratory of Knowledge Engineering with Big Data (Ministry of Education) Hefei University of Technology School of Computer Science and Information Engineering Hefei University of Technology Hefei China Department of Computer Science University of Georgia Athens GA Bio-Computing Research Center Harbin Institute of Technology (Shenzhen) Shenzhen China School of Information Science and Technology University of Science and Technology of China Hefei China

In this paper, we propose a robust representation learning model called Adaptive Structure-constrained Low-Rank Coding (AS-LRC) for the latent representation of data. To recover the underlying subspaces more accurately, AS-LRC seamlessly integrates an adaptive weighting based block-diagonal structure-constrained low-rank representation and the group sparse salient feature extraction into a unified framework. Specifically, AS-LRC performs the latent decomposition of given data into a low-rank reconstruction by a block-diagonal codes matrix, a group sparse locality-adaptive salient feature part and a sparse error part. To enforce the block-diagonal structures adaptive to different real datasets for the low-rank recovery, AS-LRC clearly computes an auto-weighting matrix based on the locality-adaptive features and multiplies by the low-rank coefficients for direct minimization at the same time. This encourages the codes to be block-diagonal and can avoid the tricky issue of choosing optimal neighborhood size or kernel width for the weight assignment, suffered in most local geometrical structures-preserving low-rank coding methods. In addition, our AS-LRC selects the L2, 1-norm on the projection for extracting group sparse features rather than learning low-rank features by Nuclear-norm regularization, which can make learnt features robust to noise and outliers in samples, and can also make the feature coding process efficient. Extensive visualizations and numerical results demonstrate the effectiveness of our AS-LRC for image representation and recovery.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Single Object Tracking for Dynamic Programming based on Fine-grained Network

引用

Journal of Physics: Conference Series 2020年第1期1576卷

作者： Jianguo Jiang Kaiyuan Jin Jingjing Wu Meibin Qi Qian Wang Key Laboratory of Knowledge Engineering with Big Data (Hefei University of Technology) Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology Hefei 230601 China

Single target tracking has always been a key and challenging research field in computer vision. Currently, an increasing number of researchers are focusing on extracting better tracking features and designing the best tracker. This paper proposes a new single target tracking network that uses fine-grained features and dynamic programming (DPFNet). In order to extract superior features, we added an attention module to the regression network enabling us to extract finer-grained and discriminative features to achieve regression. Besides, we did observe that different objects have varying moving rates; for different moving targets, the magnitude of the changes in target position within two adjacent frames is not the same either. Although an area search of 4 times the target's size can be applied to most objects, targets with large position changes may appear in other image areas outside the search area and the target would not be located as a result. Aiming at solving this problem, when designing the tracker, this paper analyzes some of the indicators for predicting the location and uses the analysis results to determine whether the search area is appropriate, so as to dynamically adjust the extent of the search area thereby significantly improving the tracking function. In other words, the size of the search area can be dynamically recalibrated for different images. Subsequent experiments prove that the method put forward in this paper achieves State-of-the-Art results.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Extremely Short Chinese Text Classification Method Based on Bidirectional Semantic Extension

引用

Journal of Physics: Conference Series 2020年第1期1437卷

作者： Yongzeng Yue Yuhong Zhang Xuegang Hu Peipei Li Key Laboratory of knowledge Engineering with Big Data (Hefei University of Technology) Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology Hefei 23061 China

Short text classification methods have achieved significant progress and wide application on text data such as Twitter and Weibo. However, the extremely short chinese texts like tax invoice data are different with traditional short texts in lackness of contextual semantic information, feature sparseness and extremely short length. The existing short text classification methods are difficult to achieve a satisfactory performance in these texts. To address these problems, this paper proposes a text classification method based on bidirectional semantic extension for extremely short texts like Chinese tax invoice data. More specifically, firstly, the Chinese knowledge graph is introduced for extending bidirectional semantic of texts and label data to expand the extremely short texts and ease the problem of feature sparseness; secondly, the hash vectorization is used to avoid the semantic problem caused by the lackness of contextual information. Experimental results conducted the real tax invoice dataset demonstrate the effectiveness of our proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Discriminative Additive Scale Loss for Deep Imbalanced Classification and Embedding

Discriminative Additive Scale Loss for Deep Imbalanced Class...

引用

IEEE International Conference on data Mining (ICDM)

作者： Zhao Zhang Weiming Jiang Yang Wang Qiaolin Ye Mingbo Zhao Mingliang Xu Meng Wang 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 AI team Shanghai Shizhuang Information Technology Co. Ltd Shanghai China College of Information Science and Technology Nanjing Forestry University Nanjing China School of Information Science and Technology Donghua University Shanghai China School of Information Engineering Zhengzhou University Zhengzhou China

ISBN: (纸本)9781665423991

Real-world data in emerging applications may suffer from highly-skewed class imbalanced distribution, however how to deal with this kind of problem appropriately through deep learning needs further investigation. In this paper, we mainly propose a novel cross-entropy based loss function, referred to as Additive Scale Loss (ASL), for deep representation learning and imbalanced classification. To deal with the class imbalanced problem, ASL aims at increasing the loss in case of misclassification, which can avoid the superimposed loss values caused by the large amount of easily classified data in the unbalanced database to dominate the loss value of misclassified data. Moreover, in real-world applications, one data source may be used for multiple scenarios, such as classification and embedding learning, however training two separable models to handle these problems is costly, especially in deep learning area. To tackle this issue, we present and integrate a discriminative inter-class separation term into ASL, and propose a discriminative ASL (D-ASL), which can not only improve the classification performance, but also obtain discriminative representations simultaneously. The discriminative inter-class separation term is general, and can be easily integrated to other loss functions, such as CE and FL, as the byproducts. Finally, a new deep convolutional neural network equipped with D-ASL and a fully-connected (FC) layer is proposed, which can classify the imbalanced image data and obtain the discriminative representations at the same time. Extensive experimental results verified the superior performance of our method.

关键词： Deep learning Training Representation learning Additives databases Soft sensors Conferences

来源：评论

学校读者我要写书评

暂无评论

Generative AI for Advanced UAV Networking

引用

IEEE Network 2024年

作者： Sun, Geng Xie, Wenwen Niyato, Dusit Du, Hongyang Kang, Jiawen Wu, Jing Sun, Sumei Zhang, Ping Jilin University College of Computer Science and Technology Changchun130012 China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Changchun130012 China Nanyang Technological University College of Computing and Data Science Singapore639798 Singapore The University of Hong Kong Department of Electrical and Electronic Engineering Hong Kong999077 Hong Kong Guangdong University of Technology School of Automation Guangzhou510006 China Agency for Science Technology and Research Institute for Infocomm Research Singapore639798 Singapore Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing100876 China

With the impressive achievements of chatGPT and Sora, generative artificial intelligence (GAI) has received increasing attention. Not limited to the field of content generation, GAI is also widely used to solve the problems in wireless communication scenarios due to its powerful learning and generalization capabilities. Therefore, we discuss key applications of GAI in improving unmanned aerial vehicle (UAV) communication and networking performance in this article. Specifically, we first review the key technologies of GAI and the important roles of UAV networking. Then, we show how GAI can improve the communication, networking, and security performances of UAV systems. Subsequently, we propose a novel framework of GAI for advanced UAV networking, and then present a case study of UAV-enabled spectrum map estimation and transmission rate optimization based on the proposed framework to verify the effectiveness of GAI-enabled UAV systems. Finally, we discuss some important open directions. © 2024 IEEE.

关键词： Aircraft communication

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：