Search

Your search keyword '"Wu, Xuanyu"' showing total 50 results
Start Over Author "Wu, Xuanyu"
50 results on '"Wu, Xuanyu"'

5. Practical Assessment of Generalization Performance Robustness for Deep Networks via Contrastive Examples

Author

Wu, Xuanyu, Li, Xuhong, Xiong, Haoyi, Zhang, Xiao, Huang, Siyu, and Dou, Dejing

Subjects
Computer Science - Machine Learning, Computer Science - Computer Vision and Pattern Recognition
Abstract

Training images with data transformations have been suggested as contrastive examples to complement the testing set for generalization performance evaluation of deep neural networks (DNNs). In this work, we propose a practical framework ContRE (The word "contre" means "against" or "versus" in French.) that uses Contrastive examples for DNN geneRalization performance Estimation. Specifically, ContRE follows the assumption in contrastive learning that robust DNN models with good generalization performance are capable of extracting a consistent set of features and making consistent predictions from the same image under varying data transformations. Incorporating with a set of randomized strategies for well-designed data transformations over the training set, ContRE adopts classification errors and Fisher ratios on the generated contrastive examples to assess and analyze the generalization performance of deep models in complement with a testing set. To show the effectiveness and the efficiency of ContRE, extensive experiments have been done using various DNN models on three open source benchmark datasets with thorough ablation studies and applicability analyses. Our experiment results confirm that (1) behaviors of deep models on contrastive examples are strongly correlated to what on the testing set, and (2) ContRE is a robust measure of generalization performance complementing to the testing set in various settings.

Published
2021

11. Interpretable Deep Learning: Interpretation, Interpretability, Trustworthiness, and Beyond

Author

Li, Xuhong, Xiong, Haoyi, Li, Xingjian, Wu, Xuanyu, Zhang, Xiao, Liu, Ji, Bian, Jiang, and Dou, Dejing

Subjects
Computer Science - Machine Learning
Abstract

Deep neural networks have been well-known for their superb handling of various machine learning and artificial intelligence tasks. However, due to their over-parameterized black-box nature, it is often difficult to understand the prediction results of deep models. In recent years, many interpretation tools have been proposed to explain or reveal how deep models make decisions. In this paper, we review this line of research and try to make a comprehensive survey. Specifically, we first introduce and clarify two basic concepts -- interpretations and interpretability -- that people usually get confused about. To address the research efforts in interpretations, we elaborate the designs of a number of interpretation algorithms, from different perspectives, by proposing a new taxonomy. Then, to understand the interpretation results, we also survey the performance metrics for evaluating interpretation algorithms. Further, we summarize the current works in evaluating models' interpretability using "trustworthy" interpretation algorithms. Finally, we review and discuss the connections between deep models' interpretations and other factors, such as adversarial robustness and learning from interpretations, and we introduce several open-source libraries for interpretation algorithms and evaluation approaches.

Published
2021

15. Evaluation of the Guidelines and Consensuses on Enuresis in Children Using the AGREEⅡand RIGHT Checklists

Author

SUN Xiangjuan, XIAO Xiang, WU Xuanyu, ZHANG Fenghua

Subjects
nocturnal enuresis, child, guidebook, consensus, agreeⅱ, right, evaluation study, research report, Medicine
Abstract

Background Enuresis in children is a common pediatric disease that can significantly reduce the quality of life of children and their families. High-quality guidelines on enuresis in children contribute to the standardization of control and improvement of clinical symptoms. So developing relevant high-quality clinical practice guidelines is a critical way to improve the diagnosis and treatment of enuresis in children. Objective To evaluate the quality of guidelines and expert consensuses on enuresis in children issued since 2010 using the AGREE Ⅱ and RIGHT checklists, aiming at providing a reference for clinical practice and future development of relevant guidelines. Methods Guidelines and expert consensuses related to enuresis in children published from January 1, 2010 to January 31, 2022 were searched in databases of CNKI, Wanfang Data, VIP, CBM, and PubMed (with searching databases of MedLive, WHO and NICE as a supplement) . The AGREE Ⅱ and RIGHT checklists were used to evaluate the methodological and reporting quality of included studies. According to the AGREEⅡ, the recommended grade of each guideline/consensus was rated as "recommended (A) ", "recommended after update (B) ", and "not recommended (C) ". The intra-class correlation coefficient (ICC) was used for the consistency test. Results Altogether, eight guidelines and five consensuses were included, which provide 185 recommendations, including 44 for diagnosis and evaluation, 140 for treatment, and one for follow-up. By the AGREEⅡchecklist, these 13 studies were assessed with average scores of (66.99%±16.73%) , (46.26%±21.56%) , (34.42%±26.73%) , (60.36% ±32.13%) , (39.82%±20.14%) and (45.35%±37.93%) for scope and purpose, stakeholder involvement, rigour of development, clarity of presentation, applicability, and editorial independence, respectively. None of the studies were with recommendation grade A, eight were with recommendation grade B, and five were with grade C. The reporting rates of the studies in accordance with seven domains of basic information, background, evidence, recommendations, review and quality assurance, funding and declaration and management of interests and other information were (60.58%±12.19%) , (50.72%±20.18%) , (30.77%±26.88%) , (32.97%±27.53%) , 25.00% (40.63%) , (25.96%±24.00%) and 16.67% (33.33%) , respectively. The guidelines had higher score in terms of the scope and purpose domain of AGREEⅡ than consensuses (P0.05) . The ICC for the reliability of each of the domains of the AGREEⅡchecklist in assessment was greater than 0.750, showing a high consistency. The ICC for the reliability of each of the domains of the RIGHT checklist was 0.736, which showed a good consistency. Conclusion The overall methodological and reporting quality of the included guidelines and consensuses needs to be improved. The recommended grades of the included literature consist of B and C without A, suggesting that high-quality guidelines should be referred first when making clinical decisions. Currently, there is no guideline for the diagnosis and treatment of enuresis in children in primary care in China, and the quality of existing guideline on Traditional Chinese Medicine management of enuresis in children needs to be improved. It is suggested to develop relevant guidelines according to the AGREE Ⅱ and RIGHT checklists to guide the clinical management of enuresis in children.

Published
2023
Full Text
View/download PDF

22. Auricular acupressure for adverse events following immunization after COVID-19 vaccine injection: A multicentre, blinded, randomized controlled trial

Author

Fu, Qinwei, Xie, Hui, Zhou, Li, Li, Xinrong, Liu, Yang, Luo, Hongyan, Zhang, Chunyan, Peng, Wenyu, Wang, Zhiqiao, Su, Chang, Xiao, Zhiyong, Lin, Hanwen, Xiao, Xiang, Wu, Xuanyu, Huang, Jiali, Wang, Xiaocen, Hu, Sihan, Tang, Jinfan, Xiao, Huan, Zhou, Jing, Feng, Chengzhi, Wang, Li, Ao, Zhimin, Chen, Xi, Zhang, Qinxiu, and Jiang, Luyun

Published
2022
Full Text
View/download PDF

28. Traditional Chinese medicine auricular point acupressure for the relief of pain, fatigue, and gastrointestinal adverse reactions after the injection of novel coronavirus-19 vaccines: a structured summary of a study protocol for a multicentre, three-arm, single-blind, prospective randomized controlled trial

Author

Fu, Qinwei, Xie, Hui, Zhou, Li, Li, Xinrong, Liu, Yang, Liu, Min, Wang, Chaoyu, Wang, Xiaocen, Wang, Zhiqiao, Tang, Jinfan, Xiao, Huan, Xiao, Zhiyong, Zhou, Jing, Feng, Chengzhi, Wang, Li, Ao, Zhimin, Chen, Xi, Su, Chang, Wu, Xuanyu, Zhao, Maolan, Hu, Sihan, Lin, Hanwen, Huang, Jiali, Xu, Guo, Zhang, Qinxiu, and Jiang, Luyun

Published
2021
Full Text
View/download PDF

29. Auricular acupressure for adverse events following immunization related to COVID-19 vaccine injection: study protocol for a multicenter, three-arm, blinded randomized controlled trial

Author

Fu, Qinwei, Xie, Hui, Zhou, Li, Li, Xinrong, Liu, Yang, Liu, Min, Wang, Chaoyu, Wang, Xiaocen, Wang, Zhiqiao, Tang, Jinfan, Xiao, Huan, Xiao, Zhiyong, Zhou, Jing, Feng, Chengzhi, Wang, Li, Ao, Zhimin, Chen, Xi, Su, Chang, Wu, Xuanyu, Zhao, Maolan, Hu, Sihan, Lin, Hanwen, Huang, Jiali, Xu, Guo, Zhang, Qinxiu, and Jiang, Luyun

Published
2021
Full Text
View/download PDF

37. Complex characteristics analysis of time-delay digital supply chain driven by cybersecurity.

Author

Dai, Daoming, Wu, Xuanyu, Si, Fengshan, Feng, Zhenan, and Chu, Weishen

Subjects
*SUPPLY chains, *INTERNET security, *GAME theory, *EVOLUTIONARY models, *DECISION making
Abstract

Purpose: The purpose of this study is to analyze the short-term development pattern and long-term development trend of the digital supply chain. Design/methodology/approach: This study uses the combination of short-term game and long-term evolutionary game theory. Findings: Findings of this study suggest that irrational decisions can make the evolutionary path of the digital supply chain complex and unpredictable. Originality/value: This study proposes an evolutionary game model for the digital supply chain that can provide good guidance for the digitalization process of enterprises. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

41. Auricular Acupressure for Adverse Events Following Immunization Related To COVID-19 Vaccine Injection: Study Protocol for a Multicenter, Three-Arm, Blinded Randomized Controlled Trial

Author

Fu, Qinwei, primary, Xie, Hui, additional, Zhou, Li, additional, Li, Xinrong, additional, Liu, Yang, additional, Liu, Min, additional, Wang, Chaoyu, additional, Wang, Xiaocen, additional, Wang, Zhiqiao, additional, Tang, Jinfan, additional, Xiao, Huan, additional, Xiao, Zhiyong, additional, Zhou, Jing, additional, Feng, Chengzhi, additional, Wang, Li, additional, Ao, Zhimin, additional, Chen, Xi, additional, Su, Chang, additional, Wu, Xuanyu, additional, Zhao, Maolan, additional, Hu, Sihan, additional, Lin, Hanwen, additional, Huang, Jiali, additional, Xu, Guo, additional, Zhang, Qinxiu, additional, and Jiang, Luyun, additional

Published
2021
Full Text
View/download PDF

42. Synthetic Biology Knowledge System

Author

Mante, Jeanet, primary, Hao, Yikai, additional, Jett, Jacob, additional, Joshi, Udayan, additional, Keating, Kevin, additional, Lu, Xiang, additional, Nakum, Gaurav, additional, Rodriguez, Nicholas E., additional, Tang, Jiawei, additional, Terry, Logan, additional, Wu, Xuanyu, additional, Yu, Eric, additional, Downie, J. Stephen, additional, McInnes, Bridget T., additional, Nguyen, Mai H., additional, Sepulvado, Brandon, additional, Young, Eric M., additional, and Myers, Chris J., additional

Published
2021
Full Text
View/download PDF

47. Scenic Slope Retaining Contrast Effect Analysis

Author

Wu Xuanyu and Zhang Shmin

Subjects
Cultural heritage, Engineering, business.industry, Engineering support, Beautification, Strength reduction, business, Retaining wall, Factor method, Civil engineering, Virescence, Finite element method
Abstract

As a natural and cultural heritage in the world, the slope of Longjing Road lies in the West Lake Scenic Area. However, this slope suffers from rain-wash, air-slake and later-on-natural movements which cause collapse, thus the slope calls for engineering support. Yet the slopes we used to build are mainly aimed to construction engineering and highways particularly, that we must take their special locations into consideration. Their particularity demand that we should choose a certainly workable engineering support and design it in a way that delights people's eye at the meantime, therefore much more difficult work is to be done to meet it. Among those engineering support we know, retaining wall or anchor-type lattice beam is helpful in large-area virescence. Through rational design and beautification, it ensures the security and stability of the engineering support and meanwhile it helps with its pleasing outlook. By means of finite element strength reduction factor method, we analyze the stability of the slope on the Longjing Road with the finite element software and the effectiveness of retaining wall and anchor-type lattice beam. Also, through safety factors, displacement and destruction process, we compare the actual effectiveness of these two cases and work out some improvement measures. These results can serve as references in choosing engineering support schemes alike.

Published
2015

Catalog

Books, media, physical & digital resources
See catalog results