Filters








27,036 Hits in 5.4 sec

Multiview Community Discovery Algorithm via Nonnegative Factorization Matrix in Heterogeneous Networks

Wang Tao, Liu Yang
2017 Mathematical Problems in Engineering  
bipartite graphs, and makes all the classifiers tend to reach a consensus on the clustering results of the target-mode nodes.  ...  The experimental results show that Joint-NMF algorithm is efficient and well-behaved in real-world heterogeneous networks and can better explore the community structure of multimode nodes in heterogeneous  ...  Objective Function of Multiview Learning via Joint-NMF. For the original NMF framework, it just considers the intertype information of 1-mode nodes.  ... 
doi:10.1155/2017/8596893 fatcat:tbn2z2z5zrevrboxxoztjg32lu

JANE: Jointly Adversarial Network Embedding

Liang Yang, Yuexue Wang, Junhua Gu, Chuan Wang, Xiaochun Cao, Yuanfang Guo
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
Extensive experiments demonstrate the remarkable superiority of the proposed JANE on link prediction (3% gains in both AUC and AP) and node clustering (5% gain in F1 score).  ...  To overcome this vital issue, a novel Joint Adversarial Network Embedding (JANE) framework is proposed to jointly distinguish the real and fake combinations of the embeddings, topology information and  ...  Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  ... 
doi:10.24963/ijcai.2020/192 dblp:conf/ijcai/0002WGWCG20 fatcat:wkckudi2wbdatczdos2hlophou

Unifying Node Labels, Features, and Distances for Deep Network Completion

Qiang Wei, Guangmin Hu
2021 Entropy  
Specifically, we first construct an estimated network topology for the unobserved part using node labels, then jointly refine the network topology and learn the edge likelihood with node labels, node features  ...  Collected network data are often incomplete, with both missing nodes and missing edges. Thus, network completion that infers the unobserved part of the network is essential for downstream tasks.  ...  Furthermore, joint node clustering and similarity learning (JCSL) [23] handles the situation, where the node features may be partially missing by computing the node similarities at the cluster level,  ... 
doi:10.3390/e23060771 pmid:34207438 pmcid:PMC8234573 fatcat:thbdhvw3b5dobeubjh5jmjknci

Learning Robot Structure and Motion Embeddings using Graph Neural Networks [article]

J. Taery Kim, Jeongeun Park, Sungjoon Choi, Sehoon Ha
2021 arXiv   pre-print
To this end, our work aims to learn embeddings for two types of robotic data: the robot's design structure, such as links, joints, and their relationships, and the motion data, such as kinematic joint  ...  We also study a few design choices of the learning framework, such as network architectures and message passing schemes.  ...  Pose embedding As shown in Fig. 6b , pose embeddings are in a single cluster and correlate to the node feature (i.e., joint angle), not the number of joints.  ... 
arXiv:2109.07543v1 fatcat:45bgxxgfzjajzcizp6mwtdodtm

User Profile Preserving Social Network Embedding

Daokun Zhang, Jie Yin, Xingquan Zhu, Chengqi Zhang
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
in node classification and clustering tasks.  ...  The theme of UPP-SNE is to embed user profile information via a nonlinear mapping into a consistent subspace, where network structure is seamlessly encoded to jointly learn informative node representations  ...  LANE [Huang et al., 2017] proposes to learn three types of latent node representations via spectral techniques from the node content-level similarity matrix, network adjacent matrix and node label-level  ... 
doi:10.24963/ijcai.2017/472 dblp:conf/ijcai/ZhangYZZ17 fatcat:ztmxzsrpobf77ieem32hd2cmpa

When Does Self-Supervision Help Graph Convolutional Networks? [article]

Yuning You, Tianlong Chen, Zhangyang Wang, Yang Shen
2020 arXiv   pre-print
Self-supervision as an emerging technique has been employed to train convolutional neural networks (CNNs) for more transferrable, generalizable, and robust representation learning of images.  ...  We first elaborate three mechanisms to incorporate self-supervision into GCNs, analyze the limitations of pretraining & finetuning and self-training, and proceed to focus on multi-task learning.  ...  .], and a US Army Research Office Young Investigator Award [W911NF2010240 to Z.W.].  ... 
arXiv:2006.09136v4 fatcat:iefaimgvxjbsxk4juymdhc5erm

A deep learning approach for uncovering lung cancer immunome patterns [article]

Moritz Hess, Stefan Lenz, Harald Binder
2018 bioRxiv   pre-print
We show that the hidden nodes of the trained networks cannot only be linked to clinical characteristics but also to specific genes, which are the visible nodes of the network.  ...  We also propose a sampling-based approach that smooths the original data according to a trained DBM and can be used for visualization and clustering.  ...  In one setting, 156 similar to the approach described in Hess et al. [5] , pre-training was performed within 157 completely separated partitions (called "partDBM no joint" in the following).  ... 
doi:10.1101/291047 fatcat:foyj3vuok5dctfoav3b4zfl6wy

Joint Association Graph Screening and Decomposition for Large-Scale Linear Dynamical Systems

Yiyuan She, Yuejia He, Shijie Li, Dapeng Wu
2015 IEEE Transactions on Signal Processing  
Based on the notion of joint association graph (JAG), we develop a joint graphical screening and estimation (JGSE) framework for efficient network learning in big data.  ...  In particular, our method can pre-determine and remove unnecessary edges based on the joint graphical structure, referred to as JAG screening, and can decompose a large network into smaller subnetworks  ...  With FLOG introduced, the two-stage JGSE learning framework is complete.  ... 
doi:10.1109/tsp.2014.2373315 fatcat:a2l2hfpwrng7vhwwnlb62h3apa

A Study of Joint Graph Inference and Forecasting [article]

Daniel Zügner, François-Xavier Aubet, Victor Garcia Satorras, Tim Januschowski, Stephan Günnemann, Jan Gasthaus
2021 arXiv   pre-print
Further, we perform ablations to study their behavior under changing conditions, e.g., when disabling the graph-learning modules and providing the ground-truth relations instead.  ...  We study a recent class of models which uses graph neural networks (GNNs) to improve forecasting in multivariate time series.  ...  Graph learning. Similar to MTGNN, GDN infers the graph by learning a node embedding v i per node.  ... 
arXiv:2109.04979v1 fatcat:pnifwirukjdgxlhhia4klkx5aa

Variational Bayes in Private Settings (VIPS) (Extended Abstract)

James R. Foulds, Mijung Park, Kamalika Chaudhuri, Max Welling
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
learning.  ...  In the full paper we extend our method to a broad class of models, including Bayesian logistic regression and sigmoid belief networks.  ...  In the above latter two settings, cuts and clustering are also typical combinatorial problems, which pose further challenge for a joint solution.  ... 
doi:10.24963/ijcai.2020/694 dblp:conf/ijcai/YanYH20 fatcat:pc4nelo7gzfmvmsiym3ohwspxa

A Survey of Community Detection Approaches: From Statistical Modeling to Deep Learning [article]

Di Jin, Zhizhi Yu, Pengfei Jiao, Shirui Pan, Dongxiao He, Jia Wu, Philip S. Yu, Weixiong Zhang
2021 arXiv   pre-print
model and deep learning.  ...  utilize deep learning and convert networked data into low dimensional representation.  ...  ., similarity-based clusters (SimClusters), which settles a multitude of recommendation tasks at Twitter via detecting bipartite communities from the useruser network and leveraging them as a representation  ... 
arXiv:2101.01669v3 fatcat:p2lkjuslmzd6hc6kpum3sz5xwq

Hypersphere ART and ARTMAP for unsupervised and supervised, incremental learning

G.C. Anagnostopoulos, M. Georgiopulos
2000 Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium  
Among these properties is fast, stable, incremental learning on the training set and good generalization on the testing set.  ...  While H-ART is intended for clustering tasks, its extension, H-ARTMAP is playing the role of Fuzzy-ARTMAP's counterpart for the supervised learning of real-valued, multi-dimensional mappings.  ...  With Fuzzy-ART and Fuzzy-ARTMAP fast learning completes in a finite number of list presentations.  ... 
doi:10.1109/ijcnn.2000.859373 dblp:conf/ijcnn/AnagnostopoulosG00 fatcat:5hgbntajbvaltd3tgh3juciebe

A Comprehensive Survey on Community Detection with Deep Learning [article]

Xing Su, Shan Xue, Fanzhen Liu, Jia Wu, Jian Yang, Chuan Zhou, Wenbin Hu, Cecile Paris, Surya Nepal, Di Jin, Quan Z. Sheng, Philip S. Yu
2021 arXiv   pre-print
Despite the classical spectral clustering and statistical inference methods, we notice a significant development of deep learning techniques for community detection in recent years with their advantages  ...  This survey devises and proposes a new taxonomy covering different state-of-the-art methods, including deep learning-based models upon deep neural networks, deep nonnegative matrix factorization and deep  ...  A Deep Neural network-based Clustering-oriented network network embedding (DNC) [113] is the extension of DNGR which joint learns node embeddings and cluster assignments.  ... 
arXiv:2105.12584v2 fatcat:matipshxnzcdloygrcrwx2sxr4

Learning Global Properties of Nonredundant Kinematic Mappings

David DeMers, Kenneth Kreutz-Delgado
1998 The international journal of robotics research  
data can be analyzed by clustering methods in order to determine the number and location of the solution branches.  ...  As a practical consequence, the inverse kinematic mapping can be directly approximated by applying neural network or other learning-based methods to each branch separately. R 3) (Pieper 1968 ).  ...  If there are M nodes in the network and N data points (N > M), the Kohonen learning algorithm adjusts their location in the data space such that each node is the nearest node to approximately N/M data  ... 
doi:10.1177/027836499801700506 fatcat:t45zx3lftzezdbqmnjkmmx5ypa

Joint Cluster Analysis of Attribute Data and Relationship Data: the Connectedk-Center Problem [chapter]

Martin Ester, Rong Ge, Byron J. Gao, Zengjian Hu, Boaz Ben-Moshe
2006 Proceedings of the 2006 SIAM International Conference on Data Mining  
We analyze the complexity of this problem and prove its NP-completeness.  ...  It is also common to observe both data types carry orthogonal information such as in market segmentation and community identification, which calls for a joint cluster analysis of both data types so as  ...  Binay Bhattacharya and Dr. Petra Berenbrink for the valuable discussions in the early stage of this study.  ... 
doi:10.1137/1.9781611972764.22 dblp:conf/sdm/EsterGGHB06 fatcat:xfgwq6vdjzdjhc7vpgoldxskoe
« Previous Showing results 1 — 15 out of 27,036 results