检索结果-内蒙古大学图书馆

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Yu Mao Zhiyi Cai Yulin Li Chunyu Shi School of Computer Science Minnan Normal University Zhangzhou China Lab of Data Science and Intelligence Application Minnan Normal University Zhangzhou China

label distribution learning (LDL) is a new learning paradigm proposed in the field of machine learning to solve the problem of label ambiguity. Many existing label distribution learning algorithms only consider the simple relationship between labels and do not mine out the complex relationship between labels. To address this problem, this paper proposes a label distribution learning based on a two-stage model error repair method(LDL-TERM). First, the initial model is used to predict the training set, and its prediction results are used as additional Features are integrated into the LDL-TER model; secondly, norm regularization is used to control the relative importance between the additional feature set and the original feature set. Experiments on 14 related data sets prove that the algorithm has better performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-label Feature Selection Based on Improved Fisher Score with label Correlation

Multi-Label Feature Selection Based on Improved Fisher Score...

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Yu Mao Yuxuan Cheng Chunyu Shi Yaojin Lin School of Computer Science Minnan Normal University Zhangzhou China Lab of Data Science and Intelligence Application Minnan Normal University Zhangzhou China

Feature selection for multi-label classification has received extensive attention in the fields of machine learning and data mining. However, some feature selection methods fail to incorporate label correlations, resulting in the degradation of learning performance. In this paper, we dedicate to fuse the Fisher score model and neighborhood mutual information model for multi-label learning. The neighborhood mutual information is initially constructed to identify the coefficients between labels, by measuring the granularity of each instance. After that, an improved Fisher score model based on label coefficients is formed to select features with multiple crucial labels. Extensive experiments demonstrate the effectiveness and stability of the proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CoCGAN: Contrastive Learning for Adversarial Category Text Generation 29

CoCGAN: Contrastive Learning for Adversarial Category Text G...

引用

29th International Conference on Computational Linguistics, COLING 2022

作者： Sheng, Xin Xu, Linli Xu, Yinlong Bao, Changcun Chen, Huang Ren, Bo Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China China School of Computer Science and Technology University of Science and Technology of China China State Key Laboratory of Cognitive Intelligence China Tencent Youtu Lab China

The task of generating texts of different categories has attracted more and more attention in the area of natural language generation recently. Meanwhile, generative adversarial net (GAN) has demonstrated its effectiveness on text generation, and is further applied to category text generation in later works. Different from existing methods, which mainly consider the pairwise relations between the text embedding and the corresponding fixed one-hot class label (data-to-class relations), this paper proposes a novel Contrastive Category Generative Adversarial Net (CoCGAN) to incorporate contrastive learning into adversarial category text generation, considering more flexible data-to-class relations as well as relations between the multiple text embeddings in the same batch (data-to-data relations). The discriminator of CoCGAN discriminates the authenticity of given samples and optimizes a contrastive learning objective to capture both more flexible data-to-class relations and data-to-data relations among training samples. Accordingly, the generator tries to produce more realistic samples which can confuse the discriminator. Experimental results on both synthetic and real category text generation datasets demonstrate that CoCGAN can achieve significant improvements over the baseline category text generation models. © 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Learning Subpocket Prototypes for Generalizable Structure-based Drug Design

arXiv

引用

arXiv 2023年

作者： Zhang, Zaixi Liu, Qi Anhui Province Key Lab of Big Data Analysis and Application University of Science and Technology of China China State Key Laboratory of Cognitive Intelligence China

Generating molecules with high binding affinities to target proteins (a.k.a. structure-based drug design) is a fundamental and challenging task in drug discovery. Recently, deep generative models have achieved remarkable success in generating 3D molecules conditioned on the protein pocket. However, most existing methods consider molecular generation for protein pockets independently while neglecting the underlying connections such as subpocket-level similarities. Subpockets are the local protein environments of lig- and fragments and pockets with similar subpockets may bind the same molecular fragment (motif) even though their overall structures are different. Therefore, the trained models can hardly generalize to unseen protein pockets in real-world applications. In this paper, we propose a novel method DrugGPS for generalizable structure-based drug design. With the biochemical priors, we propose to learn subpocket prototypes and construct a global interaction graph to model the interactions between subpocket prototypes and molecular motifs. Moreover, a hierarchical graph transformer encoder and motif-based 3D molecule generation scheme are used to improve the model’s performance. The experimental results show that our model consistently outperforms baselines in generating realistic drug candidates with high affinities in challenging out-of-distribution settings. © 2023, CC BY-NC-SA.

关键词： Proteins

来源：评论

学校读者我要写书评

暂无评论

FEDGT: FEDERATED NODE CLASSIFICATION WITH SCAlabLE GRAPH TRANSFORMER

arXiv

引用

arXiv 2024年

作者： Zhang, Zaixi Hu, Qingyong Yu, Yang Gao, Weibo Liu, Qi Anhui Province Key Lab of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China China State Key Laboratory of Cognitive Intelligence Anhui Hefei China Hong Kong University of Science and Technology Hong Kong

Graphs are widely used to model relational data. As graphs are getting larger and larger in real-world scenarios, there is a trend to store and compute subgraphs in multiple local systems. For example, recently proposed subgraph federated learning methods train Graph Neural Networks (GNNs) distributively on local subgraphs and aggregate GNN parameters with a central server. However, existing methods have the following limitations: (1) The links between local subgraphs are missing in subgraph federated learning. This could severely damage the performance of GNNs that follow message-passing paradigms to update node/edge features. (2) Most existing methods overlook the subgraph heterogeneity issue, brought by subgraphs being from different parts of the whole graph. To address the aforementioned challenges, we propose a scalable Federated Graph Transformer (FedGT) in the paper. Firstly, we design a hybrid attention scheme to reduce the complexity of the Graph Transformer to linear while ensuring a global receptive field with theoretical bounds. Specifically, each node attends to the sampled local neighbors and a set of curated global nodes to learn both local and global information and be robust to missing links. The global nodes are dynamically updated during training with an online clustering algorithm to capture the data distribution of the corresponding local subgraph. Secondly, FedGT computes clients' similarity based on the aligned global nodes with optimal transport. The similarity is then used to perform weighted averaging for personalized aggregation, which well addresses the data heterogeneity problem. Finally, extensive experimental results on 6 datasets and 2 subgraph settings demonstrate the superiority of FedGT. © 2024, CC BY.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

OFHR: Online Streaming Feature Selection With Hierarchical Structure Based on Relief 11

OFHR: Online Streaming Feature Selection With Hierarchical S...

引用

11th International Conference on Information Technology in Medicine and Education, ITME 2021

作者： Wang, Chenxi Zhang, Xiaoqing Chen, Jinkun Mao, Yu Li, Shaozi Lin, Yaojin Minnan Normal University School of Computer Science Lab of Data Science and Intelligence Application Zhangzhou China Minnan Normal University School of Mathematics and Statistics Zhangzhou China Xiamen University Department of Artificial Intelligence Xiamen China

ISBN: (纸本)9781665406796

Hierarchical classification learning, an emerging classification task in machine learning, is an essential topic. In which various feature selection algorithms have been proposed to select informative features for hierarchical classification. How-ever, existing hierarchical feature selection algorithms consider that the feature space of data is completely obtained in advance, and neglect the uncertainty and dynamism, i.e., feature arrives dynamically in an online manner. In this paper, we present an online streaming feature selection framework with hierarchical structure. First, we apply the closeness matrix between internal nodes to the Relief algorithm, which can calculate the weights of the dynamic features. Second, significant features are dynamically selected for each internal node by considering the hierarchical relationships and feature weights between nodes in the tree structure. Moreover, we perform redundant analysis of features by calculating the covariance between features, and then obtain a superior online feature subset for each internal node. Finally, the proposed algorithm is compared with six online streaming feature selection methods on six hierarchical data sets. The experimental results prove that our algorithm can improve the classification accuracy of the classifier by 10% compared to the suboptimal algorithms, which indicates that the algorithm outperforms other comparative algorithms in hierarchical data sets. © 2021 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Resource Reservation Contract Design for Mobile Edge Computing with Energy Harvesting 23

Resource Reservation Contract Design for Mobile Edge Computi...

引用

Proceedings of the 4th International Conference on Artificial intelligence and Computer Engineering

作者： Deyue Jiang Liming Chen Ziqiong Lin Wenjie Zhang Yifeng Zheng Key lab of data science and intelligence application School of Computer Sciences Minnan Normal University China Zhangzhou Affiliated Hospital of Fujian Medical University China School of Computer Sciences Minnan Normal University China

ISBN: (纸本)9798400708831

To address the bottleneck of energy performance, this paper considers an energy harvesting (EH)-enabled Mobile Edge Computing (MEC) system with contract users (CUs) and random users (RUs). The resource demands of CUs are known in advance, while RUs are mobile and randomly arriving. We propose a resource reservation framework based on contract theory with the goal of maximizing the utility of agent node (AN) under information asymmetry and demand uncertainty. First, for symmetric information scenario, the relationship between the optimal resource reservation and the demands of CUs is obtained. Then, for asymmetric information scenario, a contract-based resource reservation scheme is proposed to motivate wireless access point (WAP) to share the demand information of CUs and take part of over-reservation risk cost. Based on this, the theoretical expression for the optimal reserved resource is obtained by exploring the features of the feasible contracts under individual rationality and incentive compatibility constraints. The effectiveness of proposed contract-based resource reservation scheme is validated through simulation results.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generalized-Extended-State-Observer and Equivalent-Input-Disturbance Methods for Active Disturbance Rejection: Deep Observation and Comparison

引用

IEEE/CAA Journal of Automatica Sinica 2023年第4期10卷 957-968页

作者： Jinhua She Kou Miyamoto Qing-Long Han Min Wu Hiroshi Hashimoto Qing-Guo Wang School of Engineering Tokyo University of TechnologyHachiojiTokyo 192-0982Japan K.Miyamoto is with the Institute of Technology Shimizu CorporationKotoTokyo 135-0044Japan School of Science Computing and Engineering TechnologiesSwinburne University of TechnologyMelbourneVIC 3122Australia School of Automation China University of GeosciencesWuhan 430074 Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of EducationWuhan 430074China School of Industrial Technology Advanced Institute of Industrial TechnologyTokyo 140-0011Japan Institute of Artificial Intelligence and Future Networks Beijing Normal UniversityZhuhai 519087 Guangdong Key Lab of AI and Multi-Modal Data Processing Guangdong Provincial Key Laboratory of Interdisciplinary Research and Application for Data Science BNUHKBU United International College Zhuhai 519087China

Active disturbance-rejection methods are effective in estimating and rejecting disturbances in both transient and steady-state *** paper presents a deep observation on and a comparison between two of those methods:the generalized extended-state observer(GESO)and the equivalent input disturbance(EID)from assumptions,system configurations,stability conditions,system design,disturbance-rejection performance,and extensibility.A time-domain index is introduced to assess the disturbance-rejection performance.A detailed observation of disturbance-suppression mechanisms reveals the superiority of the EID approach over the GESO method.A comparison between these two methods shows that assumptions on disturbances are more practical and the adjustment of disturbance-rejection performance is easier for the EID approach than for the GESO method.

关键词： Active disturbance-rejection control(ADRC) disturbance observer(DOB) equivalent input disturbance(EID) extendedstate observer(ESO) generalized extended-state observer(GESO)

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Streaming Feature Selection Based on FDAF-score

Hierarchical Streaming Feature Selection Based on FDAF-score

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Zhuoxin He Yu Mao Yixiang Zeng Xiehua Yu Yaojin Lin School of Computer Science Lab of Data Science and Intelligence Application Minnan Normal University Zhangzhou China School of Computer and Information Minnan Science and Technology University Quanzhou China

In many practical classification scenarios, the data label space has a hierarchical structure, these data often have high-dimensional features, and there are irrelevant and redundant features. Thus, a variety of hierarchical classification feature selection methods are proposed. However, most of the existing algorithms ignore the temporal uncertainty and unknown of the feature space, and can only deal with static feature space data. Therefore, a new hierarchical streaming feature selection method is proposed. First, the algorithm uses the sibling strategy and the exclusion strategy in the hierarchical relationship to consider the relationship between homogeneous and heterogeneous samples. Then, the FDAF-score algorithm is integrated into the hierarchical classification streaming feature selection and used as a criterion for evaluating feature importance. Second, the improved Gaussian kernel function is used to judge the degree of redundancy between features, and the redundant features are removed. Finally, experiments on six hierarchical classification data sets show the effectiveness of the proposed algorithm.

关键词： Uncertainty Redundancy Education Feature extraction Classification algorithms Kernel Information technology

来源：评论

学校读者我要写书评

暂无评论

Backdoor Defense via Deconfounded Representation Learning

Backdoor Defense via Deconfounded Representation Learning

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Zaixi Zhang Qi Liu Zhicai Wang Zepu Lu Qingyong Hu Anhui Province Key Lab of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China State Key Laboratory of Cognitive Intelligence Hefei Anhui China University of Science and Technology of China Hong Kong University of Science and Technology

Deep neural networks (DNNs) are recently shown to be vulnerable to backdoor attacks, where attackers embed hidden backdoors in the DNN model by injecting a few poisoned examples into the training dataset. While extensive efforts have been made to detect and remove backdoors from backdoored DNNs, it is still not clear whether a backdoor-free clean model can be directly obtained from poisoned datasets. In this paper, we first construct a causal graph to model the generation process of poisoned data and find that the backdoor attack acts as the confounder, which brings spurious associations between the input images and target labels, making the model predictions less reliable. Inspired by the causal understanding, we propose the Causality-inspired Backdoor Defense (CBD), to learn deconfounded representations for reliable classification. Specifically, a backdoored model is intentionally trained to capture the confounding effects. The other clean model dedicates to capturing the desired causal effects by minimizing the mutual information with the confounding representations from the backdoored model and employing a sample-wise re-weighting scheme. Extensive experiments on multiple benchmark datasets against 6 state-of-the-art attacks verify that our proposed defense method is effective in reducing backdoor threats while maintaining high accuracy in predicting benign samples. Further analysis shows that CBD can also resist potential adaptive attacks. The code is available at https://***/zaixizhang/CBD.

关键词：

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：