Filters








321,140 Hits in 4.1 sec

Embeddings and Representation Learning for Structured Data

Benjamin Paaßen and Claudio Gallicchio and Alessio Micheli and Alessandro Sperduti
2019 arXiv   pre-print
In this contribution, we provide an high-level overview of the state-of-the-art in representation learning and embeddings for structured data across a wide range of machine learning fields.  ...  and recurrent decoder networks for structured data.  ...  Background In this section, we introduce terms that are shared among all methods for representation learning and embeddings for structured data. We begin by defining structured data itself.  ... 
arXiv:1905.06147v1 fatcat:ku5r7fuwnjdfbakiwsrzxppf3u

Unsupervised Hierarchical Graph Representation Learning by Mutual Information Maximization [article]

Fei Ding, Xiaohong Zhang, Justin Sybrandt, Ilya Safro
2020 arXiv   pre-print
Our method focuses on maximizing mutual information between "local" and high-level "global" representations, which enables us to learn the node embeddings and graph embeddings without any labeled data.  ...  The hierarchical aggregation also enables the graph representations to be explainable. In addition, supervised graph representation learning requires labeled data, which is expensive and error-prone.  ...  For graph-structured data, the learned low-dimensional representations (embeddings) can encode information of a graph's nodes, or the entire graph in the case of the GNN model.  ... 
arXiv:2003.08420v3 fatcat:n2u3becfmzdkfmyei7q3qm5xoi

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 propose a learning framework to find the representation of a robot's kinematic structure and motion embedding spaces using graph neural networks (GNN).  ...  We visualized the learned embeddings via t-SNE dimensional reduction and analyzed how the embedding space formed for different structures and pose data.  ... 
arXiv:2109.07543v1 fatcat:45bgxxgfzjajzcizp6mwtdodtm

Network representation learning method embedding linear and nonlinear network structures

Hu Zhang, Jingjing Zhou, Ru Li, Yue Fan, Mehwish Alam, Davide Buscaldi, Michael Cochez, Francesco Osborne, Diego Reforgiato Recupero, Harald Sack
2022 Semantic Web Journal  
With the rapid development of neural networks, much attention has been focused on network embedding for complex network data, which aims to learn low-dimensional embedding of nodes in the network and how  ...  Thus, the nodal characteristics of nonlinear and linear structures are explored in this paper, and an unsupervised representation method based on HGCN that joins learning of shallow and deep models is  ...  For instance, traditional feature extraction methods for machine learning mainly include data cleaning, missing value processing, data annotation and a series of manual works.  ... 
doi:10.3233/sw-212968 fatcat:trmtlpbqp5dvha7mniks3yqn44

Adversarial Network Embedding [article]

Quanyu Dai, Qiang Li, Jian Tang, Dan Wang
2017 arXiv   pre-print
., a structure preserving component and an adversarial learning component.  ...  However, except for objectives to capture network structural properties, most of them suffer from lack of additional constraints for enhancing the robustness of representations.  ...  Liang Zhang of Data Science Lab at JD.com and Prof. Xiaoming Wu of The Hong Kong Polytechnic University for their valuable discussion. Dan Wang's work is supported in part by HK PolyU G-YBAG.  ... 
arXiv:1711.07838v1 fatcat:urnmryjidfgr3di4znm6dh2p2u

Network Representation Learning: From Traditional Feature Learning to Deep Learning

Ke Sun, Lei Wang, Bo Xu, Wenhong Zhao, Shyh Wei Teng, Feng Xia
2020 IEEE Access  
., social networks, various types of network data take a great challenge for mining network data. Most representation learning algorithms just consider network structure.  ...  The transductive graph embedding is applied for predicting class label and graph context based on the input feature of observed labeled data and embeddings extracted from graph structure.  ... 
doi:10.1109/access.2020.3037118 fatcat:kca6htfarjdjpmtwcvbsppfzui

GraphFederator: Federated Visual Analysis for Multi-party Graphs [article]

Dongming Han, Wei Chen, Rusheng Pan, Yijing Liu, Jiehui Zhou, Ying Xu, Tianye Zhang, Changjie Fan, Jianrong Tao, Xiaolong Zhang
2020 arXiv   pre-print
The new federation framework consists of a shared module that is responsible for joint modeling and analysis, and a set of local modules that run on respective graph data.  ...  We also design multiple visualization views for joint visualization, exploration, and analysis of multi-party graphs.  ...  Three components are employed for constructing different types of graph representations: embedding representation, attribute representation, and structure representation.  ... 
arXiv:2008.11989v1 fatcat:oolbkzck6fhsneozw44xznve5m

Unsupervised Domain Adaptation with Feature Embeddings [article]

Yi Yang, Jacob Eisenstein
2015 arXiv   pre-print
Representation learning is the dominant technique for unsupervised domain adaptation, but existing approaches often require the specification of "pivot features" that generalize across domains, which are  ...  We show that a novel but simple feature embedding approach provides better performance, by exploiting the feature template structure common in NLP problems.  ...  We present FEMA (Feature EMbeddings for domain Adaptation), a novel representation learning approach for domain adaptation in structured feature spaces.  ... 
arXiv:1412.4385v3 fatcat:6a32k4sjzrbhdndbgf25p5fmka

Network representation learning systematic review: Ancestors and current development state

Amina Amara, Mohamed Ali Hadj Taieb, Mohamed Ben Aouicha
2021 Machine Learning with Applications  
Thus, we introduce a brief history of representation learning and word representation learning ancestor of network embedding.  ...  Artificial intelligence and machine learning have been recently leveraged as powerful systems to learn insights from network data and deal with presented challenges.  ...  to create a new data representation for machine learning based tasks.  ... 
doi:10.1016/j.mlwa.2021.100130 fatcat:axhg2gxkzfds3icebro6hlman4

Joint Embedding Learning of Educational Knowledge Graphs [article]

Siyu Yao, Ruijie Wang, Shen Sun, Derui Bu, Jun Liu
2019 arXiv   pre-print
Our model considers both structural and literal information and jointly learns embedding representations.  ...  In this paper, we focus on this problem and propose a novel model for embedding learning of educational knowledge graphs.  ...  For an educational KG which consists of triples and literals of the entities and relations, we first learn structural embedding representations based on TransE and literal embedding representations based  ... 
arXiv:1911.08776v2 fatcat:23gvecmwf5gzhdxmui4ry5tbzu

Network Embedding with Deep Metric Learning

Xiaotao CHENG, Lixin JI, Ruiyang HUANG, Ruifei CUI
2019 IEICE transactions on information and systems  
Network embedding is an important method to learn low-dimensional representations of vertices in networks, aiming to capture and preserve the network structure.  ...  The experimental results show that our method outperforms the baselines in all three datasets, and the method has proved to be effective and robust. key words: deep metric learning, network representation  ...  Acknowledgments We acknowledge Hongxin Zhi and Zhengming Liu for their inspirations. This work was partially supported by the  ... 
doi:10.1587/transinf.2018edp7233 fatcat:d54z2psz55hyxdiqdinilvdqeu

Graph Neural Networks to Advance Anticancer Drug Design [chapter]

Asmaa Rassil, Hiba Chougrad, Hamid Zouaki
2020 IFIP Advances in Information and Communication Technology  
In this work, we investigate the representation power of Node2vec for embedding learning over graphs, by comparing it to the theoretical framework Graph Isomorphism Network (GIN).  ...  We then exert the two models Node2vec and GIN to extract regular representations from chemical compounds and make predictions about their activity against lung and ovarian cancer.  ...  Graph Embedding Graph embedding frameworks [1, 3, 6, 19, 21, 26] are unsupervised approaches for learning on graph structured data.  ... 
doi:10.1007/978-3-030-49161-1_19 fatcat:skzl5c3ebfe5phmcankyxn2rgq

Deep Embedded K-Means Clustering [article]

Wengang Guo, Kaiyan Lin, Wei Ye
2021 arXiv   pre-print
In this paper, we propose DEKM (for Deep Embedded K-Means) to answer these two questions.  ...  To this end, we discard the decoder and propose a greedy method to optimize the representation. Representation learning and clustering are alternately optimized by DEKM.  ...  ACKNOWLEDGMENT The authors would like to thank anonymous reviewers for their constructive and helpful comments. This work was  ... 
arXiv:2109.15149v1 fatcat:crgavg27kvc6dlh7rddteylq3e

From Statistical Relational to Neuro-Symbolic Artificial Intelligence

Luc de Raedt, Sebastijan Dumančić, Robin Manhaeve, Giuseppe Marra
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
Neuro-symbolic and statistical relational artificial intelligence both integrate frameworks for learning with logical reasoning.  ...  These cannot only be used to characterize and position neuro-symbolic artificial intelligence approaches but also to identify a number of directions for further research.  ...  The main obstacle for neural representation learning on networks is the transformation of non-Euclidean graph structures into Euclidean embedding space, as there exists a gap between graph data and neural  ... 
doi:10.24963/ijcai.2020/677 dblp:conf/ijcai/DongHWS020 fatcat:srd5r66dovefrpa5drxmrek25e

A Self-supervised Representation Learning of Sentence Structure for Authorship Attribution [article]

Fereshteh Jafariakinabad, Kien A. Hua
2022 arXiv   pre-print
In this paper, we propose a self-supervised framework for learning structural representations of sentences.  ...  We evaluate the learned structural representations of sentences using different probing tasks, and subsequently utilize them in the authorship attribution task.  ...  The fact that the training data comes from the wide range of genres maximizes the potential efficacy for learning diverse sentence structures.  ... 
arXiv:2010.06786v2 fatcat:h3zlsaqnnbdijonbqwh7erwkbu
« Previous Showing results 1 — 15 out of 321,140 results