检索结果-内蒙古大学图书馆

IEEE International Conference on Big knowledge (ICBK)

作者： Ru Wang Meng Wu Shengwei Ji Key Laboratory of Knowledge Engineering with Big Data Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology Hefei China

ISBN: (纸本)9781665438599

knowledge Graph (KG) is a directed heterogeneous information network that contains a large number of entities and relations, which is widely used as effective side information in rec-ommender systems. Moreover, in recommender systems, the Graph Convolutional Network (GCN) model is introduced to mine the relatedness between entities in a KG because of its efficiency in extracting spatial features on topological graphs. The knowledge Graph Convolutional Network (KGCN) model up-dates the embedding of a currently positioned entity by aggregating the information of adjacent entities selected randomly. Never-theless, it has two limititations: 1) the information of neighbors se-lected randomly cannot accurately represent the current entity in the KG; 2) the model is hard to converge as graph features (i.e. The spatial relation features and semantic information features of en-tities in the KG) grow. To solve these limitations, in this paper, a meta-path (i.e., a sequence of artificially constructed relationships) is introduced into the selection of neighbors in the KGCN model to enhance the representation of each entity. Furthermore, two construction methods of the meta-path - constructing a meta-path based on the same relation (KGCN-SP) and the characteris-tics of KG (KGCN-MP) -are proposed. The experiments based on three real-world datasets demonstrate that the neighbor selection based on the meta-path is able to collect more accurate infor-mation from a KG and improve the recommendation performance effectively.

关键词： knowledge engineering Adaptation models Computational modeling Semantics Diversity reception Feature extraction data models

来源：评论

学校读者我要写书评

暂无评论

BERT-INT: A BERT-based interaction model for knowledge graph alignment 29

BERT-INT: A BERT-based interaction model for knowledge graph...

引用

29th International Joint Conference on Artificial Intelligence, IJCAI 2020

作者： Tang, Xiaobin Zhang, Jing Chen, Bo Yang, Yang Chen, Hong Li, Cuiping Key Laboratory of Data Engineering and Knowledge Engineering Ministry of Education Renmin University of China China Information School Renmin University of China China Zhejiang University China

ISBN: (纸本)9780999241165

knowledge graph alignment aims to link equivalent entities across different knowledge graphs. To utilize both the graph structures and the side information such as name, description and attributes, most of the works propagate the side information especially names through linked entities by graph neural networks. However, due to the heterogeneity of different knowledge graphs, the alignment accuracy will be suffered from aggregating different neighbors. This work presents an interaction model to only leverage the side information. Instead of aggregating neighbors, we compute the interactions between neighbors which can capture fine-grained matches of neighbors. Similarly, the interactions of attributes are also modeled. Experimental results show that our model significantly outperforms the best state-of-the-art methods by 1.9-9.7% in terms of HitRatio@1 on the dataset DBP15K. © 2020 Inst. Sci. inf., Univ. Defence in Belgrade. All rights reserved.

关键词： knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Underwater mixed spatial attention network

Underwater mixed spatial attention network

引用

2021 International Conference on Advanced Algorithms and Signal Image Processing, AASIP 2021

作者： Pan, Qiang Yang, Yueting Yang, Chuansheng Wang, Chao Tan, Anhui School of Information Engineering Zhejiang Ocean University Zhejiang Zhoushan316022 China School of Mathematics and Statistics Beihua University Jilin Jilin132013 China Key Laboratory of Oceanographic Big Data Mining and Application of Zhejiang Province Zhejiang Zhoushan316022 China

High-quality underwater images play an important role in obtaining and understanding underwater information. However, raw underwater images have many problems, such as low contrast, chromatic aberration, blur and low light, which seriously restrict the progress of other underwater tasks. In this paper, we propose an Underwater Mixed Spatial Attention Network (UMSAN) for underwater image enhancement. The evaluation on the EUVP dataset and some other real underwater images demonstrates that our network is sufficient against several of the advanced models. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Two-Stage Fusion Model for Heavy Rain Removal on Single Image

Two-Stage Fusion Model for Heavy Rain Removal on Single Imag...

引用

2021 International Conference on Advanced Algorithms and Signal Image Processing, AASIP 2021

作者： Pan, Xiaomiao Yang, Yueting Yang, Chuansheng Wang, Chao Tan, Anhui School of Information Engineering Zhejiang Ocean University Zhejiang Zhoushan316022 China School of Mathematics and Statistics Beihua University Jilin Jilin132013 China Key Laboratory of Oceanographic Big Data Mining and Application of Zhejiang Province Zhejiang Zhoushan316022 China

Current single image derain methods cannot solve the heavy rain situation well. In this paper, based on the physical model of a rainy image, we build a two-stage network, TSF-Net, which combines model-driven and data-driven methods. The first stage gets the rain streaks, atmospheric light, and transmission map to obtain the coarse rain-free image by the physical model. The second stage is a fully convolutional neural network with the structure of U-Net. The proposed Multi-Scale Projection Fusion Block (MSPFB) module, which can perceive the spatial information across the scale of images, is employed to remove the residual rain and fuzzy parts in the first stage and obtain a refined clean image. Extensive experiments show that our TSF-Net achieves better accuracy and visual improvements against state-of-the-art methods. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Rain

来源：评论

学校读者我要写书评

暂无评论

The Further Development of Intelligent knowledge for Wisdom

引用

Procedia Computer Science 2022年 214卷 1382-1388页

作者： Jifa Gu Xiao Xiao Lingling Zhang Academy of Mathematics and Systems Science of Chinese Academy of Sciences Beijing 100190 China School of Economics and Management University of Chinese Academy of Sciences Beijing 100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing 100190 China MOE Social Science Laboratory of Digital Economic Forecasts and Policy Simulation at UCAS Beijing 100190 China

This paper introduces some modes from D, I, K to W, M. From different angles we investigate the various process. Especially we utilize the intelligent knowledge to develop the whole process. Finally we design a new medical hall for workshop of meta-synthetic engineering for helping patient to get the appropriate treatment and useful knowledge for health.

关键词： Wisdom Intelligent knowledge DIKWM Medical Hall for Workshop of Meta-synthetic engineering

来源：评论

学校读者我要写书评

暂无评论

ReconBoost: boosting can achieve modality reconcilement 24

ReconBoost: boosting can achieve modality reconcilement

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Cong Hua Qianqian Xu Shilong Bao Zhiyong Yang Qingming Huang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China and School of Cyber Security University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China and Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores a novel multi-modal alternating learning paradigm pursuing a reconciliation between the exploitation of uni-modal features and the exploration of cross-modal interactions. This is motivated by the fact that current paradigms of multi-modal learning tend to explore multi-modal features simultaneously. The resulting gradient prohibits further exploitation of the features in the weak modality, leading to modality competition, where the dominant modality overpowers the learning process. To address this issue, we study the modality-alternating learning paradigm to achieve reconcilement. Specifically, we propose a new method called ReconBoost to update a fixed modality each time. Herein, the learning objective is dynamically adjusted with a reconcilement regularization against competition with the historical models. By choosing a KL-based reconcilement, we show that the proposed method resembles Friedman's Gradient-Boosting (GB) algorithm, where the updated learner can correct errors made by others and help enhance the overall performance. The major difference with the classic GB is that we only preserve the newest model for each modality to avoid overfitting caused by ensembling strong learners. Furthermore, we propose a memory consolidation scheme and a global rectification scheme to make this strategy more effective. Experiments over six multi-modal benchmarks speak to the efficacy of the method. We release the code at https://***/huacong/ReconBoost.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sampling topic representative users by integrating node degree and edge weight 5

Sampling topic representative users by integrating node degr...

引用

5th IEEE International Conference on data Science in Cyberspace, DSC 2020

作者： Wang, Jiangwen Ban, Zhijie Inner Mongolia A.R. Key Laboratory of Data Mining and Knowledge Engineering College of Computer Science Inner Mongolia University Huhhot China China Banking and Insurance Information Technology Management Co. Ltd Beijing China

ISBN: (纸本)9781728195582

Finding a subset of users to statistically represent the original social network in terms of attributes has received a lot of attention recently. The existing literature had proved that the problem is NP-Hard and proposed some sampling models, but representative degrees aren't systematically studied. Taking node degree and edge weight into account, we present a topic sampling model for the whole social network in this paper. Firstly, by statistically calculating node degrees for one real dataset, we find that one node is more likely to be a representative user if its degree is high. We then give some definitions about representative degrees. Based on statistical stratified sample, we finally give a sampling model of topic representative users who are from most or all attribute groups in the network. Experimental results show that our algorithm significantly outperforms other six baseline methods and is also effective in terms of time complexity. © 2020 IEEE.

关键词： Sampling

来源：评论

学校读者我要写书评

暂无评论

Generating Action-conditioned Prompts for Open-vocabulary Video Action Recognition 24

Generating Action-conditioned Prompts for Open-vocabulary Vi...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Jia, Chengyou Luo, Minnan Chang, Xiaojun Dang, Zhuohang Han, Mingfei Wang, Mengmeng Dai, Guang Dang, Sizhe Wang, Jingdong School of Computer Science and Technology MOEKLINNS Lab Xi'an Jiaotong University Shaanxi Xi'an China University of Science and Technology of China Anhui Hefei China School of Computer Science and Technology Xi'an Jiaotong University Shaanxi Xi'an China ReLER Lab AAII University of Technology Sydney SydneyNSW Australia Zhejiang University of Technology College of Computer Science and Technology China SGIT AI Lab State Grid Corporation of China Beijing China Baidu Inc Beijing China United Arab Emirates Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi'an Jiaotong University Xi'an710049 China SGIT AI Lab State Grid Corporation of China China School of Computer Science and Technology Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi'an Jiaotong University Xi'an710049 China

ISBN: (纸本)9798400706868

Exploring open-vocabulary video action recognition is a promising venture, which aims to recognize previously unseen actions within any arbitrary set of categories. Existing methods typically adapt pretrained image-text models to the video domain, capitalizing on their inherent strengths in generalization. A common thread among such methods is the augmentation of visual embeddings with temporal information to improve the recognition of seen actions. Yet, they compromise with standard less-informative action descriptions, thus faltering when confronted with novel actions. Drawing inspiration from human cognitive processes, we argue that augmenting text embeddings with human prior knowledge is pivotal for open-vocabulary video action recognition. To realize this, we innovatively blend video models with Large Language Models (LLMs) to devise Action-conditioned Prompts. Specifically, we harness the knowledge in LLMs to produce a set of descriptive sentences that contain distinctive features for identifying given actions. Building upon this foundation, we further introduce a multi-modal action knowledge alignment mechanism to align concepts in video and textual knowledge encapsulated within the prompts. Extensive experiments on various video benchmarks, including zero-shot, few-shot, and base-to-novel generalization settings, demonstrate that our method not only sets new SOTA performance but also possesses excellent interpretability. © 2024 ACM.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

NMR Spectra Denoising with Vandermonde Constraints

arXiv

引用

arXiv 2023年

作者： Guo, Di Xu, Runmin Wu, Jinyu Lin, Meijin Du, Xiaofeng Qu, Xiaobo School of Computer and Information Engineering Fujian Engineering Research Center for Medical Data Mining and Application Xiamen University of Technology Xiamen361024 China Department of Electronic Science Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance Xiamen University Xiamen361102 China Department of Applied Marine Physics and Engineering Xiamen University Xiamen361102 China

Nuclear magnetic resonance (NMR) spectroscopy serves as an important tool to analyze chemicals and proteins in bioengineering. However, NMR signals are easily contaminated by noise during the data acquisition, which can affect subsequent quantitative analysis. Therefore, denoising NMR signals has been a long-time concern. In this work, we propose an optimization model-based iterative denoising method, CHORD-V, by treating the time-domain NMR signal as damped exponentials and maintaining the exponential signal form with a Vandermonde factorization. Results on both synthetic and realistic NMR data show that CHORD-V has a superior denoising performance over typical Cadzow and rQRd methods, and the state-of-the-art CHORD method. CHORD-V restores low-intensity spectral peaks more accurately, especially when the noise is relatively high. © 2023, CC BY.

关键词： Nuclear magnetic resonance spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Disentangled Noisy Correspondence Learning

arXiv

引用

arXiv 2024年

作者： Dang, Zhuohang Luo, Minnan Wang, Jihong Jia, Chengyou Han, Haochen Wan, Herun Dai, Guang Chang, Xiaojun Wang, Jingdong The School of Computer Science and Technology The Ministry of Education Key Laboratory of Intelligent Networks and Network Security The Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi'an Jiaotong University Shaanxi Xi'An710049 China SGIT AI Lab State Grid Shaanxi Electric Power Company Limited State Grid Corporation of China Shaanxi China The School of Information Science and Technology University of Science and Technology China United Arab Emirates The Baidu Inc China

Cross-modal retrieval is crucial in understanding latent correspondences across modalities. However, existing methods implicitly assume well-matched training data, which is impractical as real-world data inevitably involves imperfect alignments, i.e., noisy correspondences. Although some works explore similarity-based strategies to address such noise, they suffer from sub-optimal similarity predictions influenced by modality-exclusive information (MEI), e.g., background noise in images and abstract definitions in texts. This issue arises as MEI is not shared across modalities, thus aligning it in training can markedly mislead similarity predictions. Moreover, although intuitive, directly applying previous cross-modal disentanglement methods suffers from limited noise tolerance and disentanglement efficacy. Inspired by the robustness of information bottlenecks against noise, we introduce DisNCL, a novel information-theoretic framework for feature Disentanglement in Noisy Correspondence Learning, to adaptively balance the extraction of MII and MEI with certifiable optimal cross-modal disentanglement efficacy. DisNCL then enhances similarity predictions in modality-invariant subspace, thereby greatly boosting similarity-based alleviation strategy for noisy correspondences. Furthermore, DisNCL introduces soft matching targets to model noisy many-to-many relationships inherent in multi-modal input for noise-robust and accurate cross-modal alignment. Extensive experiments confirm DisNCL's efficacy by 2% average recall improvement. Mutual information estimation and visualization results show that DisNCL learns meaningful MII/MEI subspaces, validating our theoretical analyses. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

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：