检索结果-内蒙古大学图书馆

International Symposium on software Reliability engineering (ISSRE)

作者： Zhihao Liu Yang Zheng Xiaoting Du Zheng Hu Wenjie Ding Yanming Miao Zheng Zheng School of Automation Science and Electrical Engineering Beihang University Beijing China School of Computer Science (National Pilot Software Engineering School)Beijing University of Posts and Telecommunications Beijing China TTE lab Huawei Shenzhen China Central Software Institute Huawei Shenzhen China

ISBN: (数字)9781665451321

ISBN: (纸本)9781665451338

Deep learning libraries are the cornerstone of deep learning systems, and millions of deep learning applications are built on top of deep learning libraries. Due to long-term continuous running, many numerical operations and heavy dependence on resources, deep learning libraries are prone to the effects of software aging. Aging in deep learning libraries can threaten the reliability of deep learning systems and make training and application of deep learning more time-consuming and expensive, causing users to lose confidence in it. In this work, we manually screened 138 bug reports containing aging-related bugs from a total of 13,694 bug reports in four popular deep learning libraries (i.e., TensorFlow, MXNET, PaddlePaddle and MindSpore). We analyzed the information in these 138 bug reports to answer three questions: What categories of aging-related bugs exist in deep learning libraries? What is the distribution of different categories of aging-related bugs in deep learning libraries? Which deep learning phases are most susceptible to software aging? Finally, we conducted a fine-grained taxonomy of aging-related bugs, including four levels and seventeen categories, and obtained eight important findings with corresponding practical implications.

关键词： Deep learning Training computer bugs Taxonomy Aging Libraries software

来源：评论

学校读者我要写书评

暂无评论

Adversarial self-training improves robustness and generalization for gradual domain adaptation 23

Adversarial self-training improves robustness and generaliza...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Lianghe Shi Weiwei Liu School of Computer Science Wuhan University and National Engineering Research Center for Multimedia Software Wuhan University and Institute of Artificial Intelligence Wuhan University and Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University

Gradual Domain Adaptation (GDA), in which the learner is provided with additional intermediate domains, has been theoretically and empirically studied in many contexts. Despite its vital role in security-critical scenarios, the adversarial robustness of the GDA model remains unexplored. In this paper, we adopt the effective gradual self-training method and replace vanilla self-training with adversarial self-training (AST). AST first predicts labels on the unlabeled data and then adversarially trains the model on the pseudo-labeled distribution. Intriguingly, we find that gradual AST improves not only adversarial accuracy but also clean accuracy on the target domain. We reveal that this is because adversarial training (AT) performs better than standard training when the pseudo-labels contain a portion of incorrect labels. Accordingly, we first present the generalization error bounds for gradual AST in a multiclass classification setting. We then use the optimal value of the Subset Sum Problem to bridge the standard error on a real distribution and the adversarial error on a pseudo-labeled distribution. The result indicates that AT may obtain a tighter bound than standard training on data with incorrect pseudo-labels. We further present an example of a conditional Gaussian distribution to provide more insights into why gradual AST can improve the clean accuracy for GDA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ear Recognition Based on Gabor-SIFT 6th

Ear Recognition Based on Gabor-SIFT

引用

6th International Conference on Artificial Intelligence and Security, ICAIS 2020

作者： Tian, Ying Dong, Huiwen Wang, Libing School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

ISBN: (纸本)9783030578831

Scale invariant feature transform is a local point features extraction method. It can find those feature vectors in different scale space which are invariant for scale changes and rotations, and are flexible for illumination variations and affine transformations. The paper chooses SIFT to extract key points of ear images. Then the features of key points are extracted with the local multi-scale analysis feature of the Gabor wavelet. In this way, every key point is represented by a series of multi-scale and multi-orientation Gabor filter coefficients. Finally Ear recognition based on these feature is carried out with Euclidean distance as similarity measurement. Experimental results show that proposed method can effectively extract ear feature points, and obtain high recognition rate by using few feature points. It is robust to rigid changes, illumination and rotations changes of ear image, provides a new approach to the research for ear recognition. © 2020, Springer Nature Switzerland AG.

关键词： Gabor filters

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Domain Adaptation with Implicit Pseudo Supervision for Semantic Segmentation

arXiv

引用

arXiv 2022年

作者： Xu, Wanyu Wang, Zengmao Bian, Wei The School of Computer Science Wuhan University Institute of Artificial Intelligence Wuhan University National Engineering Research Center for Multimedia Software Wuhan University Wuhan430072 China The School of Computer Science UTS Australia

Pseudo-labelling is a popular technique in unsupervised domain adaptation for semantic segmentation. However, pseudo labels are noisy and inevitably have confirmation bias due to the discrepancy between source and target domains and training process. In this paper, we train the model by the pseudo labels which are implicitly produced by itself to learn new complementary knowledge about target domain. Specifically, we propose a tri-learning architecture, where every two branches produce the pseudo labels to train the third one. And we align the pseudo labels based on the similarity of the probability distributions for each two branches. To further implicitly utilize the pseudo labels, we maximize the distances of features for different classes and minimize the distances for the same classes by triplet loss. Extensive experiments on GTA5 to Cityscapes and SYNTHIA to Cityscapes tasks show that the proposed method has considerable improvements. © 2022, CC BY-NC-ND.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Image and Feature Space Based Domain Adaptation for Vehicle Detection

引用

computers, Materials & Continua 2020年第12期65卷 2397-2412页

作者： Ying Tian Libing Wang Hexin Gu Lin Fan School of Computer Science and Software Engineering University of Science and Technology LiaoningAnshan114051China School of Mechanical Engineering and Automation University of Science and Technology LiaoningAnshan114051China Faculty of Business Economics&LawThe University of QueenslandBrisbaneQLD 4072Australia

The application of deep learning in the field of object detection has experienced much ***,due to the domain shift problem,applying an off-the-shelf detector to another domain leads to a significant performance drop.A large number of ground truth labels are required when using another domain to train models,demanding a large amount of human and financial *** order to avoid excessive resource requirements and performance drop caused by domain shift,this paper proposes a new domain adaptive approach to cross-domain vehicle *** approach improves the cross-domain vehicle detection model from image space and feature *** employ objectives of the generative adversarial network and cycle consistency loss for image style transfer in image *** feature space,we align feature distributions between the source domain and the target domain to improve the detection *** are carried out using the method with two different datasets,proving that this technique effectively improves the accuracy of vehicle detection in the target domain.

关键词： Deep learning cross-domain vehicle detection

来源：评论

学校读者我要写书评

暂无评论

Graph-Order Optimization Algorithm Based on Equal-in-Space Distance Model for High-Resolution Image Matting 7

Graph-Order Optimization Algorithm Based on Equal-in-Space D...

引用

7th IEEE International Conference on Cloud Computing and Intelligence Systems, CCIS 2021

作者： Feng, Fujian Huang, Han Liang, Yihui South China University of Technology School of Software Engineering Guangzhou510006 China Ministry ofEducation Guangzhou510006 China School ofData Science and Information Engineering Guizhou Minzu University Guiyang550025 China Zhongshan Institute University of Electronic Science and Technology of China School of Computer Science Zhongshan528400 China

ISBN: (纸本)9781665441490

Image matting is an essential image processing technology. optimized-based image matting methods can significantly improve the alpha matte quality of high-resolution images. However, the local information of the foreground may be similar to the background, which causes the inversion problem of the alpha matte in the single-point optimized. In this paper, we propose an image matting mathematical model of the equal-in-space distance. The model transforms the high-resolution image matting problem into several small-scale combinatorial optimization problems according to the similarity among pixel features. Inspired by spanning tree, we propose a graph-order optimization strategy, which generates the optimization sequence of small-scale optimization problems according to the edge weight among graph nodes. In addition, we designed a graph-order optimization algorithm based on optimized information transfer to solve each node in the graph. Experimental results show that the proposed model solves the alpha matte inversion problem of single-point optimization matting. Besides, the proposed algorithm outperforms the state-of-the-art optimization algorithms for the high-resolution image matting problem. © 2021 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Simple Image-level Classification Improves Open-vocabulary Object Detection

arXiv

引用

arXiv 2023年

作者： Fang, Ruohuan Pang, Guansong Bai, Xiao School of Computer Science and Engineering Beihang University China School of Computing and Information Systems Singapore Management University Singapore State Key Laboratory of Software Development Environment Jiangxi Research Institute Beihang University China

Open-Vocabulary Object Detection (OVOD) aims to detect novel objects beyond a given set of base categories on which the detection model is trained. Recent OVOD methods focus on adapting the image-level pre-trained vision-language models (VLMs), such as CLIP, to a region-level object detection task via, e.g., region-level knowledge distillation, regional prompt learning, or region-text pre-training, to expand the detection vocabulary. These methods have demonstrated remarkable performance in recognizing regional visual concepts, but they are weak in exploiting the VLMs' powerful global scene understanding ability learned from the billion-scale image-level text descriptions. This limits their capability in detecting hard objects of small, blurred, or occluded appearance from novel/base categories, whose detection heavily relies on contextual information. To address this, we propose a novel approach, namely Simple Image-level Classification for Context-Aware Detection Scoring (SIC-CADS), to leverage the superior global knowledge yielded from CLIP for complementing the current OVOD models from a global perspective. The core of SIC-CADS is a multi-modal multi-label recognition (MLR) module that learns the object co-occurrence-based contextual information from CLIP to recognize all possible object categories in the scene. These image-level MLR scores can then be utilized to refine the instance-level detection scores of the current OVOD models in detecting those hard objects. This is verified by extensive empirical results on two popular benchmarks, OV-LVIS and OV-COCO, which show that SIC-CADS achieves significant and consistent improvement when combined with different types of OVOD models. Further, SIC-CADS also improves the cross-dataset generalization ability on Objects365 and OpenImages. Code is available at https://***/mala-lab/SIC-CADS. Copyright © 2023, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Multi-Instance Active Learning for Relation Extraction

Multi-Instance Active Learning for Relation Extraction

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Meilin Cao Jiqiong Jiang Jing Zhang School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China Research Center of Intelligent Computing Software Zhejiang Lab Hangzhou China

Achieving high-accuracy relation extraction usually requires a large well-annotated corpus for model training, which is extremely time-consuming and labor-intensive. Consequently, distant supervision, implemented as a multi-instance learning paradigm, was proposed to build models from bags using automated labeled corpora, where massive noises are unavoidably introduced in the bag labels, deteriorating the model performance. To deal with the above problem, this paper proposes novel multi-instance active learning for relation extraction, which aims to improve the classification performance of the learned models under the premise of using as few labeled training samples as possible. The proposed active learning strategies for bag selection are based on the similarity between bags, the improved Fisher matrix, and the combination of the two (hybrid), respectively. Extensive experiment results on two real-world datasets under three base learners consistently show that our proposed active learning methods archive better performance for relation extraction, compared with three baseline methods. Furthermore, the hybrid bag selection strategy performs the best.

关键词： Training Learning systems

来源：评论

学校读者我要写书评

暂无评论

Detection and Replay of Distributed Denial of Service Attacks In Smart Cities Using a Hybrid Deep Learning Approach

Detection and Replay of Distributed Denial of Service Attack...

引用

Business Analytics for Technology and Security (ICBATS), International Conference on

作者： Muhammad Waseem Iqbal Ghassan F. Issa Muhammad Yousif Muhammad Atif Department of Software Engineering Superior University Lahore Pakistan School of Information Technology Skyline University College Sharjah UAE Department of Computer Science NCBAE Lahore. Pakistan Applied Science Research Center Applied Science Private University Amman Jordan Department of Computer Science NCBAE Lahore Pakistan

Infrastructure for smart cities is presently being built, mostly because of the (IoT) platform, which connects a wide variety of things by using the website's underlying infrastructure. As a result, it is possible to employ a uniform platform to automate the services that are delivered. Smart city infrastructure, on the other hand, is susceptible to cyberattacks owing to security flaws in IoT networks. Distributing denial of service (DDoS) and replay assaults, for example, violate the requirements for certification in smart cities. The lack of citations in this part backs up many of the statements it makes. Who needs a citation? Both assaults have the potential to substantially damage smart city infrastructure, which might also cost money and even result in human fatalities. This paper covers the creation of a blended deep- learning algorithm for replay and DDoS intrusion prevention on a growth-oriented smart city platform. This is made possible by combining proven machine-learning approaches with more recent advances in artificial intelligence. We simulate scattered denial of service and replay attacks on three datasets acquired from real-world smart cities to assess the efficacy of the proposed hybrid strategy on environmental, smart river, and smart soil dataset. The suggested model has shown excellent rates of accuracy. The ecosystem collection has an accuracy level of 98.37%, the smart riverbed dataset must have an accuracy of 98.13%, and the smart mud dataset must have an accuracy of 99.51%. The findings demonstrated that the proposed model outperformed past instances of evolutionary computation and machine learning techniques employed in the corpus of academic literature.

关键词： Analytical models Machine learning algorithms Smart cities Computational modeling Soil Probability distribution Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Image Reconstruction with B0 Inhomogeneity using an Interpretable Deep Unrolled Network on an Open-bore MRI-Linac

arXiv

引用

arXiv 2024年

作者： Shan, Shanshan Gao, Yang Waddington, David E.J. Chen, Hongli Whelan, Brendan Liu, Paul Z.Y. Wang, Yaohui Liu, Chunyi Gan, Hongping Gao, Mingyuan Liu, Feng Soochow University Suzhou215123 China Image X Institute Sydney School of Health Sciences Faculty of Medicine and Health University of Sydney Sydney2015 Australia Department of Medical Physics Ingham Institute of Applied Medical Research Liverpool2170 Australia School of Computer Science and Engineering Central South University Changsha410083 China School of Electrical Engineering and Computer Science University of Queensland Brisbane4072 Australia Institute of Electrical Engineering Chinese Academy of Science Beijing100190 China School of Software Northwestern Polytechnical University Taicang215400 China

MRI-Linac systems require fast image reconstruction with high geometric fidelity to localize and track tumours for radiotherapy treatments. However, B0 field inhomogeneity distortions and slow MR acquisition potentially limit the quality of the image guidance and tumour treatments. In this study, we develop an interpretable unrolled network, referred to as RebinNet, to reconstruct distortion-free images from B0 inhomogeneity-corrupted k-space for fast MRI-guided radiotherapy applications. RebinNet includes convolutional neural network (CNN) blocks to perform image regularizations and nonuniform fast Fourier Transform (NUFFT) modules to incorporate B0 inhomogeneity information. The RebinNet was trained on a publicly available MR dataset from eleven healthy volunteers for both fully sampled and subsampled acquisitions. Grid phantom and human brain images acquired from an open-bore 1T MRI-Linac scanner were used to evaluate the performance of the proposed network. The RebinNet was compared with the conventional regularization algorithm and our recently developed UnUNet method in terms of root mean squared error (RMSE), structural similarity (SSIM), residual distortions, and computation time. Imaging results demonstrated that the RebinNet reconstructed images with lowest RMSE (0.92) at four-time acceleration for simulated brain images. The RebinNet could better preserve structural details and substantially improve the computational efficiency (ten-fold faster) compared to the conventional regularization methods, and had better generalization ability than the UnUNet method. The proposed RebinNet can achieve rapid image reconstruction and overcome the B0 inhomogeneity distortions simultaneously, which would facilitate accurate and fast image guidance in radiotherapy treatments. © 2024, CC BY-NC-ND.

关键词： Magnetic resonance imaging

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：