检索结果-内蒙古大学图书馆

Joint multi-task cascade for instance segmentation

学校读者我要写书评

暂无评论

Joint multi-task cascade for instance segmentation

作者： Wen, Yaole Hu, Fuyuan Ren, Jinchang Shang, Xinru Li, Linyan Xi, Xuefeng School of Electronic & Information Engineering Suzhou University of Science and Technology SuzhouJiangsu215009 China Centre for Signal and Image Processing University of Strathclyde Glasgow United Kingdom Suzhou Institute of Trade & Commerce SuzhouJiangsu215009 China Virtual Reality Key Laboratory of Intelligent Interaction and Application Technology of Suzhou Suzhou University of Science and Technology SuzhouJiangsu215009 China Suzhou Key Laboratory for Big Data and Information Service Suzhou University of Science and Technology SuzhouJiangsu215009 China

Instance segmentation requires both pixel-level classification accuracy and high-level semantic features at the target instance level, which is very challenging, and the cascade structure can effectively improve both of these problems. To make full use of the relationship between detection and segmentation, this paper proposes a joint multi-tasking cascade structure, which is not simply to cascade the two tasks of detection and segmentation, but to unitedly put them into multi-stage processing, and especially to integrate the information at different stages of the mask branch. The entire structure can effectively utilize the superior characteristics of each stage in the matter of detection and segmentation, thus improving the quality of mask prediction. The feature fusion process is introduced in the full convolution networks (FCN) branch, and the high-level and low-level features are effectively fused to enhance the contextual information of the picture semantic features. The experiments demonstrate the better results on the COCO dataset. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Semantics

Breaking the Curse of Space Explosion: Towards Efficient NAS with Curriculum Search

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Guo, Yong Chen, Yaofo Zheng, Yin Zhao, Peilin Chen, Jian Huang, Junzhou Tan, Mingkui School of Software Engineering South China University of Technology China Pazhou Laboratory China Weixin Group Tencent China Tencent AI Lab Tencent China Guangdong Key Laboratory of Big Data Analysis and Processing China

Neural architecture search (NAS) has become an important approach to automatically find effective architectures. To cover all possible good architectures, we need to search in an extremely large search space with billions of candidate architectures. More critically, given a large search space, we may face a very challenging issue of space explosion. However, due to the limitation of computational resources, we can only sample a very small proportion of the architectures, which provides insufficient information for the training. As a result, existing methods may often produce suboptimal architectures. To alleviate this issue, we propose a curriculum search method that starts from a small search space and gradually incorporates the learned knowledge to guide the search in a large space. With the proposed search strategy, our Curriculum Neural Architecture Search (CNAS) method significantly improves the search efficiency and finds better architectures than existing NAS methods. Extensive experiments on CIFAR-10 and imageNet demonstrate the effectiveness of the proposed method. Copyright © 2020, The Authors. All rights reserved.

关键词： Curricula

Adaptive hybrid projective models for aerial video stitching 6

学校读者我要写书评

暂无评论

Adaptive hybrid projective models for aerial video stitching

6th International Conference on Systems and Informatics, ICSAI 2019

作者： Xi, Runping Han, Tianyi Zhang, Yaqing Han, Qianqian Natl. Engineering Laboratory of Integrated Aero-Space-Ground-Ocean Big Data Application Technology Shaanxi Provincial Key Laboratory of Speech Image Information Processing School of Computer Science Northwestern Polytechnical University Xi'an China

ISBN: (纸本)9781728152561

Aerial video stitching has been widely used in many fields such as large-scale videos surveillance, disaster monitoring and unmanned aerial vehicle (UAV) navigation. Ghosting and parallax are challenging problems when the video streams are captured by UAV. This paper presents a novel method based on hybrid projective model which combines local warp alignment and global warp projection for better registration accuracy. The geometric alignment model is used for global registration of images and the content-preserving alignment model is used for local refinement. Simultaneously, reference frame is updated adaptively according to the shift of scenery, which is used to eliminate the cumulative error appearing in long video streams. The proposed approach has been tested under multiple datasets in different scenarios, which shows good performance. © 2019 IEEE.

关键词： Security systems

VSRQ: Quantitative Assessment Method for Safety Risk of Vehicle Intelligent Connected System

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Tian Guan, Wenshan Miao, Hao Huang, Xiujie Liu, Zhiquan Wang, Chaonan Guan, Quanlong Fang, Liangda Duan, Zhifei Department of Cyberspace Security College of Information Science and Technology Jinan University Guangzhou511486 China Department of Computer Science College of Information Science and Technology Jinan University Guangzhou510632 China College of Information Science and Technology Jinan University Guangdong Institution of Smart Education Jinan University Guangzhou510632 China College of Cyber Security Jinan University Guangzhou510632 China Guangdong Gene Data Processing and Analysis Engineering Research Center China Pazhou Laboratory Guangzhou510330 China Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin541004 China Guangzhou XPeng Motors Technology Co. Ltd No.8 Songgang Road Changxing Street Cencun Tianhe District Guangzhou China

The field of intelligent connected in modern vehicles continues to expand, and the functions of vehicles become more and more complex with the development of the times. This has also led to an increasing number of vehicle vulnerabilities and many safety issues. Therefore, it is particularly important to identify high-risk vehicle intelligent connected systems, because it can inform security personnel which systems are most vulnerable to attacks, allowing them to conduct more thorough inspections and tests. In this paper, we develop a new model for vehicle risk assessment by combining I-FAHP with FCA clustering: VSRQ model. We extract important indicators related to vehicle safety, use fuzzy cluster analys (FCA) combined with fuzzy analytic hierarchy process (FAHP) to mine the vulnerable components of the vehicle intelligent connected system, and conduct priority testing on vulnerable components to reduce risks and ensure vehicle safety. We evaluate the model on OpenPilot and experimentally demonstrate the effectiveness of the VSRQ model in identifying the safety of vehicle intelligent connected systems. The experiment fully complies with ISO 26262 and ISO/SAE 21434 standards, and our model has a higher accuracy rate than other models. These results provide a promising new research direction for predicting the security risks of vehicle intelligent connected systems and provide typical application tasks for VSRQ. The experimental results show that the accuracy rate is 94.36%, and the recall rate is 73.43%, which is at least 14.63% higher than all other known indicators. Copyright © 2023, The Authors. All rights reserved.

关键词： Vehicle safety

Inferential text generation with multiple knowledge sources and meta-learning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Guo, Daya Asai, Akari Tang, Duyu Duan, Nan Gong, Ming Shou, Linjun Jiang, Daxin Yin, Jian Zhou, Ming School of Data and Computer Science Sun Yat-sen University Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China University of Washington Microsoft Research Asia Beijing China Microsoft Search Technology Center Asia Beijing China

We study the problem of generating inferential texts of events for a variety of commonsense like if-else relations. Existing approaches typically use limited evidence from training examples and learn for each relation individually. In this work, we use multiple knowledge sources as fuels for the model. Existing commonsense knowledge bases like ConceptNet are dominated by taxonomic knowledge (e.g., isA and relatedTo relations), having a limited number of inferential knowledge. We use not only structured commonsense knowledge bases, but also natural language snippets from search-engine results. These sources are incorporated into a generative base model via key-value memory network. In addition, we introduce a meta-learning based multi-task learning algorithm. For each targeted commonsense relation, we regard the learning of examples from other relations as the meta-training process, and the evaluation on examples from the targeted relation as the meta-test process. We conduct experiments on Event2Mind and ATOMIC datasets. Results show that both the integration of multiple knowledge sources and the use of the meta-learning algorithm improve the performance. Copyright © 2020, The Authors. All rights reserved.

关键词： Learning algorithms

学校读者我要写书评

暂无评论

International Conference on Networking and Network Applications (NaNA)

作者： Zhongmin Wang Rui Zhou Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and Telecommunications Xi'an Shaanxi China School of Computer Science and Technology Xi'an University of Posts and Telecommunications Xi'an Shaanxi China

ISBN: (数字)9781728189543

ISBN: (纸本)9781728189550

The information interaction when the brain processes emotional activities is intricate. Therefore, it is very necessary for us to explore the mechanisms of the functional coordination of various brain regions. Recent studies have demonstrated that there are hub regions in the brain that are highly connected and highly central, and are responsible for coordinating dynamic interactions between brain regions. This study aims to construct the dynamic brain networks related to emotions and examine the structure of the hubs of these networks that reflect functional connectivity. In this paper, we demonstrate that a combination of network hubs set a closely connected collective called “rich-club”,which can be used as the core network architecture for identifying different emotions. According to the phase synchronization relationship of the Electroencephalogram (EEG) in three time periods to construct the corresponding brain networks in different sub-states, we found that the dynamic connection patterns produced by different emotions are different, which is manifested in the different rich-club organization. The discovery of rich-club structure related to emotions in this study is of great significance for studying the brain mechanisms behind human emotional processes.

关键词： Emotion recognition Organizations Color Network architecture Electroencephalography Synchronization

Distance variety preserving hashing for large-scale retrieval

学校读者我要写书评

暂无评论

Distance variety preserving hashing for large-scale retrieva...

作者： Zhai, Sheping School of Computer Science and Technology Xi'an University of Posts and Telecommunications Xi'an710121 China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and Telecommunications Xi'an710121 China

Hashing techniques have been seen huge adoption in large-scale retrieval owing to the computational and storage efficiencies of binary codes. However, the binary codes might not preserve the semantic similarity powerfully and lead to poor accuracy. To evade binary limitation but remain hashing superiority, this paper introduces a novel asymmetric hashing framework for distance variety preserving, which yields preferable performance via real-valued query and multi-valued database with shorter code length. Concretely, a binary sparse representation is proposed for multivalued database encoding to make computation and storage complexity tolerable. By exploiting the supervised information, we learn a projective matrix that project data onto a sub-space in which the similarity between realvalued query and multi-valued database can be measured. Due to the discrete constraints, for different discrete variables, we present different meticulous discrete solutions for their sub-problems. Exhaustive experiments on three multilabel datasets demonstrate our approach outperforms most state-of-the-art hashing methods. © 2018 John Wiley & Sons, Ltd.

关键词： database systems

Improved drug-target interaction prediction with Intermolecular Graph Transformer

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Siyuan Wang, Yusong Wang, Tong Deng, Yifan He, Liang Shao, Bin Yin, Jian Zheng, Nanning Liu, Tie-Yan School of Computer Science and Engineering Sun Yat-Sen University Guangzhou510006 China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China Microsoft Research Asia Beijing100080 China Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an710049 China School of Computer Science Fudan University Shanghai200433 China

The identification of active binding drugs for target proteins (termed as drug-target interaction prediction) is the key challenge in virtual screening, which plays an essential role in drug discovery. Although recent deep learning-based approaches achieved better performance than molecular docking, existing models often neglect certain aspects of the intermolecular information, hindering the performance of prediction. We recognize this problem and propose a novel approach named Intermolecular Graph Transformer (IGT) that employs a dedicated attention mechanism to model intermolecular information with a three-way Transformer-based architecture. IGT outperforms state-of-the-art approaches by 9.1% and 20.5% over the second best for binding activity and binding pose prediction respectively, and shows superior generalization ability to unseen receptor proteins. Furthermore, IGT exhibits promising drug screening ability against SARS-CoV-2 by identifying 83.1% active drugs that have been validated by wet-lab experiments with near-native predicted binding poses. © 2021, CC BY-NC-SA.

关键词： Proteins

On the Detection of Shilling Attacks in Federated Collaborative Filtering

学校读者我要写书评

暂无评论

On the Detection of Shilling Attacks in Federated Collaborat...

Reliable Distributed Systems

作者： Yangfan Jiang Yipeng Zhou Di Wu Chao Li Yan Wang School of Data and Computer Science Sun Yat-sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing (Sun Yat-sen University) Ministry of Education China Macquarie University Sydney NSW Australia Tencent Inc Shenzhen China

ISBN: (数字)9781728176260

ISBN: (纸本)9781728176277

Federated collaborative filtering (Fed-CF) is a variant of federated learning (FL) models, which can protect user privacy in recommender systems. In Fed-CF, the recommendation model is collectively trained across multiple decentralized clients by exchanging gradients only. However, the decentralized nature of Fed-CF makes it vulnerable to shilling attacks, which can be realized by inserting fake ratings of target items to distort recommendation results. Unfortunately, previous detection algorithms cannot work well in the FL framework, as all original data samples are not disclosed at all. In this paper, we are the first to systematically study the problem of shilling attacks in the context of federated learning, and propose an effective detection method called Federated Shilling Attack Detector (FSAD) to detect shilling attackers in Fed-CF. We first show the feasibility of shilling attacks in Fed-CF. Next, we dedicatedly design four novel features based on exchanged gradients among clients. By incorporating these gradient-based features, we train a semi-supervised Bayes classifier to identify shilling attackers effectively. Finally, we conduct extensive experiments based on real-world datasets to evaluate the performance of our proposed FSAD method. The experimental results show that FSAD can detect shilling attackers in Fed-CF with high accuracy, with the F 1 value as high as 0.90 on the Netflix dataset, which approaches the performance of the optimal detector that utilizes complete private user information for detection.

关键词： Privacy Collaborative filtering Detectors Collaborative work Information filters Reliability Recommender systems