检索结果-内蒙古大学图书馆

PCA: Semi-Supervised Segmentation with Patch Confidence Adversarial Training

IEEE Transactions on Network Science and engineering 2025年

作者： Xu, Zhenghua Yang, Runhe Xu, Zihang Zhang, Shuo Yang, Yuchen Liu, Weipeng Xu, Weichao Chen, Junyang Lukasiewicz, Thomas Leung, Victor C. M. Hebei University of Technology State Key Laboratory of Reliability and Intelligence of Electrical Equipment School of Health Sciences and Biomedical Engineering Tianjin China The University of Hong Kong Department of Electrical and Electronic Engineering Hong Kong The Johns Hopkins University Department of Applied Mathematics and Statistic Baltimore United States Hebei University of Technology School of Artificial Intelligence Tianjin China Hebei Provincial Hospital of Traditional Chinese Medicine Digestive Disease Center Shijiazhuang China Shenzhen University College of Computer Science and Software Engineering Shenzhen China University of Oxford Department of Computer Science University of Oxford United Kingdom Department of Computer Science United Kingdom Vienna University of Technology institute of Logic and Computation nstitute of Logic and Computation Vienna Austria

Deep-learning-based semi-supervised learning (SSL) methods have achieved a strong performance in medical image segmentation, which can alleviate doctors' expensive annotation by utilizing a large amount of unlabeled data. Unlike most existing semi-supervised learning methods, adversarial training methods distinguish samples from different sources by learning the data distribution of the segmentation map, leading the segmenter to generate more accurate predictions. We argue that the current performance restrictions for such approaches are the problems of feature extraction and learning preferences. In this paper, we propose a new semi-supervised adversarial method called Patch Confidence Adversarial Training (PCA) for medical image segmentation. The PCA method's discriminator penalizes patch-level structures, guiding the generator to optimize different patch areas, by leveraging pixel context, the generator is driven to focus on high-frequency features, making it harder to deceive the discriminator and easy to converge to an ideal state, which more effectively guides the segmenter to generate high-quality pseudo-labels. Furthermore, at the discriminator's input, we supplement image information constraints, making it simpler to fit the expected data distribution. Extensive experiments on the Automated Cardiac Diagnosis Challenge (ACDC) 2017 dataset and the Brain Tumor Segmentation (BraTS) 2019 challenge dataset show that our method outperforms the state-of-the-art semi-supervised methods, which demonstrates its effectiveness for medical image segmentation. © 2025 IEEE. All rights reserved.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Adversarial Learning with Cost-Sensitive Classes

arXiv

引用

arXiv 2021年

作者： Shen, Haojing Chen, Sihong Wang, Ran Wang, Xizhao Big Data Institute College of Computer Science and Software Engineering Guangdong Key Lab. of Intelligent Information Processing Shenzhen University Guangdong Shenzhen518060 China The College of Mathematics and Statistics Shenzhen University Shenzhen518060 China The Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China

It is necessary to improve the performance of some special classes or to particularly protect them from attacks in adversarial learning. This paper proposes a framework combining cost-sensitive classification and adversarial learning together to train a model that can distinguish between protected and unprotected classes, such that the protected classes are less vulnerable to adversarial examples. We find in this framework an interesting phenomenon during the training of deep neural networks, called Min-Max property, that is, the absolute values of most parameters in the convolutional layer approach zero while the absolute values of a few parameters are significantly larger becoming bigger. Based on this Min-Max property which is formulated and analyzed in a view of random distribution, we further build a new defense model against adversarial examples for adversarial robustness improvement. An advantage of the built model is that it performs better than the standard one and can combine with adversarial training to achieve an improved performance. It is experimentally confirmed that, regarding the average accuracy of all classes, our model is almost as same as the existing models when an attack does not occur and is better than the existing models when an attack occurs. Specifically, regarding the accuracy of protected classes, the proposed model is much better than the existing models when an attack occurs. © 2021, CC BY.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Theoretical evidence of H-He demixing under Jupiter and Saturn conditions

arXiv

引用

arXiv 2023年

作者： Chang, Xiaoju Chen, Bo Zeng, Qiyu Wang, Han Chen, Kaiguo Tong, Qunchao Yu, Xiaoxiang Kang, Dongdong Zhang, Shen Guo, Fangyu Hou, Yong Zhao, Zengxiu Yao, Yansun Ma, Yanming Dai, Jiayu College of Science National University of Defense Technology Changsha410073 China Hunan Key Laboratory of Extreme Matter and Applications National University of Defense Technology Changsha410073 China Laboratory of Computational Physics Institute of Applied Physics and Computational Mathematics Beijing100088 China Department of Physics and Engineering Physics University of Saskatchewan SaskatoonSKS7N 5E2 Canada State Key Lab of Superhard Materials International Center for Computational Method and Software College of Physics Jilin University Changchun130012 China International Center of Future Science Jilin University Changchun130012 China

The immiscibility of hydrogen-helium mixture under the temperature and pressure conditions of planetary interiors is crucial for understanding the structures of gas giant planets (e.g., Jupiter and Saturn). While the experimental probe at such extreme conditions is challenging, theoretical simulation is heavily relied in an effort to unravel the mixing behavior of hydrogen and helium. Here we develop a method via a machine learning accelerated molecular dynamics simulation to quantify the physical separation of hydrogen and helium under the conditions of planetary interiors. The immiscibility line achieved with the developed method yields substantially higher demixing temperatures at pressure above 1.5 Mbar than earlier theoretical data, but matches better to the experimental estimate. Our results suggest a possibility that H-He demixing takes place in a large fraction of the interior radii of Jupiter and Saturn, i.e., 27.5% in Jupiter and 48.3% in Saturn. This indication of an H-He immiscible layer hints at the formation of helium rain and offers a potential explanation for the decrease of helium in the atmospheres of Jupiter and Saturn. © 2023, CC BY.

关键词： Planets

来源：评论

学校读者我要写书评

暂无评论

A Distributed Approximation Algorithm for the Total Dominating Set Problem

SSRN

引用

SSRN 2022年

作者： Wang, Limin Zhang, Zhao Du, Donglei Mao, Yaping Zhang, Xiaoyan State Key Laboratory for Novel Software Technology Nanjing University Jiangsu210023 China College of Mathematics and Computer Sciences Zhejiang Normal University Zhejiang321004 China Faculty of Management University of New Brunswick FrederictonNBE3B 5A3 Canada School of Mathematics and Statistics Qinghai Normal University Qinghai810008 China Academy of Plateau Science and Sustainability Qinghai Xining810008 China School of Mathematical Science Institute of Mathematics Nanjing Normal University Jiangsu210023 China

Given a network graph $G=(V,E)$, a subset $T\subseteq V$ is said to be a total dominating set (TDS) if every $v\in V$ is adjacent to at least one node in $T$. In this paper, we present a distributed algorithm for the TDS problem via the LP relaxation techniques. For a positive integer $k$ and maximum degree $\Delta$, the proposed algorithm outputs a fractional total dominating set of expected size $O(k\Delta^\frac{2}{k})$. It is the first distributed algorithm for TDS which obtains an approximation ratio in $O(k^2)$ communication rounds where the algorithm use messages of size $O(\log\Delta)$;then the fractional solution is rounded to obtain an integer total dominating set for the original problem. © 2022, The Authors. All rights reserved.

关键词： Approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials

arXiv

引用

arXiv 2025年

作者： Zeng, Jinzhe Zhang, Duo Peng, Anyang Zhang, Xiangyu He, Sensen Wang, Yan Liu, Xinzijian Bi, Hangrui Li, Yifan Cai, Chun Zhang, Chengqian Du, Yiming Zhu, Jia-Xin Mo, Pinghui Huang, Zhengtao Zeng, Qiyu Shi, Shaochen Qin, Xuejian Yu, Zhaoxi Luo, Chenxing Ding, Ye Liu, Yun-Pei Shi, Ruosong Wang, Zhenyu Bore, Sigbjørn Løland Chang, Junhan Deng, Zhe Ding, Zhaohan Han, Siyuan Jiang, Wanrun Ke, Guolin Liu, Zhaoqing Lu, Denghui Muraoka, Koki Oliaei, Hananeh Singh, Anurag Kumar Que, Haohui Xu, Weihong Xu, Zhangmancang Zhuang, Yong-Bin Dai, Jiayu Giese, Timothy J. Jia, Weile Xu, Ben York, Darrin M. Zhang, Linfeng Wang, Han School of Artificial Intelligence and Data Science Unversity of Science and Technology of China Hefei China AI for Science Institute Beijing100080 China DP Technology Beijing100080 China Academy for Advanced Interdisciplinary Studies Peking University Beijing100871 China State Key Lab of Processors Institute of Computing Technology Chinese Academy of Sciences Beijing100871 China University of Chinese Academy of Sciences Beijing China Baidu Inc. Beijing China Department of Computer Science University of Toronto TorontoON Canada Department of Chemistry Princeton University PrincetonNJ08540 United States University of Chinese Academy of Sciences Beijing100871 China State Key Laboratory of Physical Chemistry of Solid Surfaces iChEM College of Chemistry and Chemical Engineering Xiamen University Xiamen361005 China College of Integrated Circuits Hunan University Changsha410082 China State Key Laboratory of Advanced Technology for Materials Synthesis and Processing Center for Smart Materials and Device Integration School of Material Science and Engineering Wuhan University of Technology Wuhan430070 China College of Science National University of Defense Technology Changsha410073 China Hunan Key Laboratory of Extreme Matter and Applications National University of Defense Technology Changsha410073 China ByteDance Research Beijing100098 China Ningbo Institute of Materials Technology and Engineering Chinese Academy of Sciences Ningbo315201 China College of Materials Science and Opto-Electronic Technology University of Chinese Academy of Sciences Beijing100049 China Key Laboratory of Theoretical and Computational Photochemistry of Ministry of Education College of Chemistry Beijing Normal University Beijing100875 China Department of Geosciences Princeton University PrincetonNJ08544 United States Department of Applied Physics and Applied Mathematics Columbia University New YorkNY10027 United States IKKEM Fujian Xiamen361005 China Graduate

In recent years, machine learning potentials (MLPs) have become indispensable tools in physics, chemistry, and materials science, driving the development of software packages for molecular dynamics (MD) simulations and related applications. These packages, typically built on specific machine learning frameworks such as TensorFlow, PyTorch, or JAX, face integration challenges when advanced applications demand communication across different frameworks. The previous TensorFlow-based implementation of DeePMD-kit exemplified these limitations. In this work, we introduce DeePMD-kit version 3, a significant update featuring a multi-backend framework that supports TensorFlow, PyTorch, JAX, and PaddlePaddle backends, and demonstrate the versatility of this architecture through the integration of other MLPs packages and of Differentiable Molecular Force Field. This architecture allows seamless backend switching with minimal modifications, enabling users and developers to integrate DeePMD-kit with other packages using different machine learning frameworks. This innovation facilitates the development of more complex and interoperable workflows, paving the way for broader applications of MLPs in scientific research. Copyright © 2025, The Authors. All rights reserved.

关键词： Molecular dynamics

来源：评论

学校读者我要写书评

暂无评论

Monopole effects, core excitations, and β decay in the A = 130 hole nuclei near ^(132)Sn

引用

Chinese Physics C 2019年第5期43卷 35-40页

作者： Han-Kui Wang Zhi-Hong Li Cen-Xi Yuan Zhi-Qiang Chen Ning Wang Wei Qin Yi-Qi He College of Physics and Telecommunication Engineering Zhoukou Normal University China Institute of Atomic Energy Sino-French Institute of Nuclear Engineering and Technology Sun Yat-Sen University School of Physics and State Key Laboratory of Nuclear Physics and Technology Peking University College of mathematics and statistics Zhoukou Normal University

The proton and neutron cross-shell excitations across the Z = 50 shell are investigated in the southwest quadrant of ^(132) Sn by large-scale shell-model calculations with extended pairing and multipole-multipole force. The model space allows proton(neutron) core excitations, and both the low-and high-energy states for ^(130) In are well described, as found by comparison with the experimental data. The monopole effects between the proton orbit and neutron orbit are studied as the new monopole correction that perfectly reproduces the first 1^+ level in ^(130) In. The energy interval of proton(neutron) core excitations in ^(130) In lies in the range of 4.5-6.5(2.0-4.1) MeV, and the high energy yrast states are predicted as neutron core excitations. The decays are calculated among the A=130 nuclei of ^(130) In, ^(130) Sn and ^(130) Cd.

关键词： monopole effects core-excitations β decay hole nuclei

来源：评论

学校读者我要写书评

暂无评论

Evolution of Maximum Bending Strain on Poisson's Ratio Distribution

arXiv

引用

arXiv 2023年

作者： Li, Yang Zhang, Le Wang, Dehua Hou, Limei Du, Shanmei Deng, Yang Du, Yanfeng Xin, Yingfei Fu, Chongyang Gu, Yan Wang, Xiaoxiong College of Physics Science Qingdao University Qingdao266071 China School of Physics and Electronic Engineering Jiangsu Normal University Jiangsu221116 China School of Mathematics and Statistics Qingdao University Qingdao266071 China Collaborative Innovation Center for Eco-Textiles of Shandong Province State Key Laboratory of Bio-Fibers and Eco-Textiles Qingdao University Qingdao266071 China University-Industry Joint Center for Ocean Observation and Broadband Communication College of Physics Qingdao University Qingdao266071 China Weihai Innovation Research Institute of Qingdao University Weihai264200 China

In recent years, new flexible functional materials have attracted increasinginterest, but there is a lack of the designing mechanisms of flexibility designwith superstructures. In traditional engineering mechanics, the maximum bendingstrain (MBS) was considered universal for describing the bendable properties ofa given material, leading to the universal designing method of lowering thedimension such as thin membranes designed flexible functional *** thiswork, the MBS was found only applicable for materials with uniformlydistributed Poisson\'s ratio, while the MBS increases with the thickness of thegiven material in case there is a variation Poisson\'s ratio in different *** means the MBS can be enhanced by certain Poisson\'s ratio design in thefuture to achieve better flexibility of thick materials. Here, the inorganicfreestanding nanofiber membranes, which have a nonconstant Poisson\'s ratioresponse on stress/strain for creating nonuniformly distributed Poisson\'s ratiowere proven applicable for designing larger MBS and lower Young\'s modulus forthicker samples. Copyright © 2023, The Authors. All rights reserved.

关键词： Functional materials

来源：评论

学校读者我要写书评

暂无评论

Spectral CT Reconstruction via Low-rank Representation and Structure Preserving Regularization

arXiv

引用

arXiv 2022年

作者： He, Yuanwei Zeng, Li Xu, Qiong Wang, Zhe Yu, Haijun Shen, Zhaoqiang Yang, Zhaojun Zhou, Rifeng College of Mathematics and Statistics Chongqing University Chongqing401331 China Engineering Research Center of Industrial Computed Tomography Nondestructive Testing Ministry of Education Chongqing University Chongqing400044 China Beijing Engineering Research Center of Radiographic Techniques and Equipment Institute of High Energy Physics Chinese Academy of Sciences Beijing100049 China Jinan Laboratory of Applied Nuclear Science Jinan250131 China Key Lab of Optoelectronic Technology and Systems Ministry of Education Chongqing University Chongqing400044 China State Key Laboratory of Mechanical Transmission Chongqing University Chongqing400044 China

Objective: With the development of computed tomography (CT) imaging technology, it is possible to acquire multi-energy data by spectral CT. Being different from conventional CT, the X-ray energy spectrum of spectral CT is cutting into several narrow bins which leads to the result that only a part of photon can be collected in each individual energy channel, which cause the image qualities to be severely degraded by noise and artifacts. To address this problem, we propose a spectral CT reconstruction algorithm based on low-rank representation and structure preserving regularization in this paper. Approach: To make full use of the prior knowledge about both the inter-channel correlation and the sparsity in gradient domain of inner-channel data, this paper combines a low-rank correlation descriptor with a structure extraction operator as priori regularization terms for spectral CT reconstruction. Furthermore, a split-Bregman based iterative algorithm is developed to solve the reconstruction model. Finally, we propose a multi-channel adaptive parameters generation strategy according to CT values of each individual energy channel. Main results: Experimental results on numerical simulations and real mouse data indicate that the proposed algorithm achieves higher accuracy on both reconstruction and material decomposition than the methods based on simultaneous algebraic reconstruction technique (SART), total variation minimization (TVM), total variation with low-rank (LRTV), and spatial-spectral cube matching frame (SSCMF). Compared with SART, our algorithm improves the feature similarity (FSIM) by 40.4% on average for numerical simulation reconstruction, whereas TVM, LRTV, and SSCMF correspond to 26.1%, 28.2%, and 29.5%, respectively. Significance: We outline a multi-channel reconstruction algorithm tailored for spectral CT. The qualitative and quantitative comparisons present a significant improvement of image quality, indicating its promising potential in spectral CT ima

关键词： Image quality

来源：评论

学校读者我要写书评

暂无评论

BLOCKS OF THE CATEGORY Op IN TYPE E

arXiv

引用

arXiv 2020年

作者： Jieren, Hu Xiao, Wei Zhang, Ailin College of Computer Science and Software Engineering Shenzhen University ShenzhenGuangdong518060 China College of Mathematics and statistics Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen Guangdong518060 China College of Mathematics and statistics Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen Guangdong518060 China

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dense networks with scale-free feature

引用

Physical Review E 2020年第5期101卷 052317-052317页

作者： Fei Ma Xiaomin Wang Ping Wang Xudong Luo School of Electronics Engineering and Computer Science Peking University Beijing 100871 China National Engineering Research Center for Software Engineering Peking University Beijing 100871 China School of Software and Microelectronics Peking University Beijing 102600 China and Key Laboratory of High Confidence Software Technologies (PKU) Ministry of Education Beijing 100871 China College of Mathematics and Statistics Northwest Normal University Lanzhou 730070 China

While previous works have shown that an overwhelming number of scale-free networks are sparse, there still exist some real-world networks including social networks, urban networks, information networks, which are by observation dense. In this paper, we propose a framework for generating scale-free graphs with a dense feature using two simple yet helpful operations: first-order subdivision and line operation. From the theoretical point of view, our method can be used not only to produce desired scale-free graphs with a density feature, i.e., a power-law exponent γ falling into the interval 1<γ≤2, but also to establish many other unexpected networked models, for instance, power-law models having a large diameter. In addition, the networked models generated upon our framework show an especially assortative structure. That is, their own Pearson correlation coefficients are able to achieve the theoretical upper bound. Last but not the least, we find the sizes of community in the proposed models to follow the power law in a form with respect to modularity maximization.

关键词： Network structure Networks & random structures Scale free & inhomogeneous 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：