Filters








3,345 Hits in 5.9 sec

Hyperbolic Function Embedding: Learning Hierarchical Representation for Functions of Source Code in Hyperbolic Space

Mingming Lu, Yan Liu, Haifeng Li, Dingwu Tan, Xiaoxian He, Wenjie Bi, Wendbo Li
2019 Symmetry  
Considering the inherent hierarchy of functions, we propose a novel hyperbolic function embedding (HFE) method, which can learn a distributed and hierarchical representation for each function via the Poincaré  ...  Hence, learning an effective representation model for the functions of source code, from a modern view, is a crucial problem.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/sym11020254 fatcat:7orayibrf5b4tkjcsjferfhtka

Embedding Text in Hyperbolic Spaces

Bhuwan Dhingra, Christopher Shallue, Mohammad Norouzi, Andrew Dai, George Dahl
2018 Proceedings of the Twelfth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-12)  
We apply this framework to learn word and sentence embeddings in hyperbolic space in an unsupervised manner from text corpora.  ...  The learned hyperbolic embeddings show improvements over Euclidean embeddings in some -but not all -downstream tasks, suggesting that hierarchical organization is more useful for some tasks than others  ...  When learning representations of such datasets, hyperbolic spaces have recently been advocated as alternatives to the standard Euclidean spaces in order to better represent the hierarchical structure  ... 
doi:10.18653/v1/w18-1708 dblp:conf/textgraphs/DhingraSNDD18 fatcat:xmfgkg7jg5fhpnk2qv2zi6wk5q

Embedding Text in Hyperbolic Spaces [article]

Bhuwan Dhingra, Christopher J. Shallue, Mohammad Norouzi, Andrew M. Dai, George E. Dahl
2018 arXiv   pre-print
We apply this framework to learn word and sentence embeddings in hyperbolic space in an unsupervised manner from text corpora.  ...  The learned hyperbolic embeddings show improvements over Euclidean embeddings in some -- but not all -- downstream tasks, suggesting that hierarchical organization is more useful for some tasks than others  ...  When learning representations of such datasets, hyperbolic spaces have recently been advocated as alternatives to the standard Euclidean spaces in order to better represent the hierarchical structure  ... 
arXiv:1806.04313v1 fatcat:nvbydehtqncsbktgi674fqerca

Cross-lingual Word Embeddings in Hyperbolic Space [article]

Chandni Saxena, Mudit Chaudhary, Helen Meng
2022 arXiv   pre-print
Our results show that by preserving the latent hierarchical information, hyperbolic spaces can offer better representations for cross-lingual embeddings.  ...  space to learn unsupervised cross-lingual word representations from a German-English parallel corpus.  ...  ., 2018) learn a mapping function from a source embedding space to the target embedding space based on different objective criteria.  ... 
arXiv:2205.01907v1 fatcat:dxqla5zzdfbarnyp2x2e2cm5cm

Hyperbolic Embeddings for Learning Options in Hierarchical Reinforcement Learning [article]

Saket Tiwari, M. Prannoy
2019 arXiv   pre-print
We apply the recent advancements of learning embeddings using Riemannian optimisation in the hyperbolic space to embed the state set into the hyperbolic space and create a model of the environment.  ...  We demonstrate empirically, both in discrete and continuous domains, how these embeddings can improve the learning of meaningful sub-tasks.  ...  Natural gradient works efficiently in learning. Neural Comput., 10 (2)  ... 
arXiv:1812.01487v2 fatcat:cguvyaa3gfgcfhbwrqh27yadba

HEAT: Hyperbolic Embedding of Attributed Networks [article]

David McDonald, Shan He
2019 arXiv   pre-print
Finding a low dimensional representation of hierarchical, structured data described by a network remains a challenging problem in the machine learning community.  ...  To fill this gap, we introduce HEAT (Hyperbolic Embedding of ATributed networks), the first method for embedding attributed networks to a hyperbolic space.  ...  An emerging representation learning approach for complex networks is hyperbolic embedding.  ... 
arXiv:1903.03036v2 fatcat:76jvo6jgp5ejvhmakhap7v4mby

Semi-Supervised Hierarchical Drug Embedding in Hyperbolic Space [article]

Ke Yu, Shyam Visweswaran, Kayhan Batmanghelich
2020 arXiv   pre-print
The hyperbolic space is amenable for encoding hierarchical concepts.  ...  Learning accurate drug representation is essential for tasks such as computational drug repositioning and prediction of drug side-effects.  ...  The hyperbolic VAE learns an embedding for codes that are amenable to hierarchical representation.  ... 
arXiv:2006.00986v1 fatcat:zcrwigpnajfrxor6l2xjvhsj7y

Low-Dimensional Hyperbolic Knowledge Graph Embeddings [article]

Ines Chami, Adva Wolf, Da-Cheng Juan, Frederic Sala, Sujith Ravi, Christopher Ré
2020 arXiv   pre-print
For hierarchical data, hyperbolic embedding methods have shown promise for high-fidelity and parsimonious representations.  ...  KGs often exhibit hierarchical and logical patterns which must be preserved in the embedding space.  ...  Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views, policies, or endorsements, either expressed or implied  ... 
arXiv:2005.00545v1 fatcat:mm3ej3ptirh7xmrytkcuiygr6y

Hyperbolic Node Embedding for Signed Networks [article]

Wenzhuo Song, Hongxu Chen, Xueyan Liu, Hongzhe Jiang, Shengsheng Wang
2020 arXiv   pre-print
Signed network embedding methods aim to learn vector representations of nodes in signed networks.  ...  For instance, previous works did not consider the hierarchical structures of networks, which is widely witnessed in real-world networks.  ...  We propose a non-Euclidean representation learning method for signed networks named Hyperbolic Signed Network Embedding (HSNE).  ... 
arXiv:1910.13090v2 fatcat:yrqagsg3zzhnlofkkpnft7gs4a

Incorporating Domain Knowledge into Health Recommender Systems Using Hyperbolic Embeddings [chapter]

Joel Peito, Qiwei Han
2021 Studies in Computational Intelligence  
With recent advances in representation learning enabling the hierarchical embedding of health knowledge into the hyperbolic Poincare space, this work proposes a content-based recommender system for patient-doctor  ...  matchmaking in primary care based on patients' health profiles, enriched by pre-trained Poincare embeddings of the ICD-9 codes through transfer learning.  ...  Nevertheless, recent works [2, 3] proposing hyperbolic embeddings for learning hierarchical representations appear to provide a bypass for this issue.  ... 
doi:10.1007/978-3-030-65351-4_11 fatcat:z7c3no2xljcopesrumr7bjqize

Knowledge Association with Hyperbolic Knowledge Graph Embeddings [article]

Zequn Sun, Muhao Chen, Wei Hu, Chengming Wang, Jian Dai, Wei Zhang
2020 arXiv   pre-print
Recent related methods built on Euclidean embeddings are challenged by the hierarchical structures and different scales of KGs.  ...  We propose a hyperbolic relational graph neural network for KG embedding and capture knowledge associations with a hyperbolic transformation.  ...  We thank the anonymous reviewers for their insightful comments.  ... 
arXiv:2010.02162v1 fatcat:lhah4lzokzdafib2judmzzzvhu

Hyperbolic Hierarchical Knowledge Graph Embeddings for Link Prediction in Low Dimensions [article]

Wenjie Zheng, Wenxue Wang, Fulan Qian, Shu Zhao, Yanping Zhang
2022 arXiv   pre-print
For hierarchical data, instead of traditional Euclidean space, hyperbolic space as an embedding space has shown the promise of high fidelity and low memory consumption; however, existing hyperbolic KGE  ...  Experiments show that HypHKGE can effectively model semantic hierarchies in hyperbolic space and outperforms the state-of-the-art hyperbolic methods, especially in low dimensions.  ...  ACKNOWLEDGMENTS This work is jointly supported by the National Natural Science Foundation of China (No. 61876001), and the Natural  ... 
arXiv:2204.13704v1 fatcat:2o7gb7alsvbnteduuiffi2yawy

Hyperbolic Disk Embeddings for Directed Acyclic Graphs [article]

Ryota Suzuki, Ryusuke Takahama, Shun Onoda
2019 arXiv   pre-print
Tackling in this problem, we develop Disk Embeddings, which is a framework for embedding DAGs into quasi-metric spaces.  ...  Existing state-of-the-art methods, Order Embeddings and Hyperbolic Entailment Cones, are instances of Disk Embedding in Euclidean space and spheres respectively.  ...  Acknowledgements We would like to thank Tatsuya Shirakawa, Katsuhiko Hayashi, and all the anonymous reviewers for their insightful comments, advice and suggestions.  ... 
arXiv:1902.04335v3 fatcat:rdsfynet75cbfnuckpereplafy

Self-Supervised Graph Learning with Hyperbolic Embedding for Temporal Health Event Prediction [article]

Chang Lu, Chandan K. Reddy, Yue Ning
2021 arXiv   pre-print
To address these challenges, we first propose a hyperbolic embedding method with information flow to pre-train medical code representations in a hierarchical structure.  ...  Many data-driven approaches employ temporal features in EHR for predicting specific diseases, readmission times, or diagnoses of patients.  ...  We take advantage of the Poincaré ball model [28] , [29] to learn the representations of the hierarchical structure, which encodes nodes in H to a hyperbolic space.  ... 
arXiv:2106.04751v1 fatcat:pumxem2aqfdwbmk5t4qnxzfibu

Word2Box: Learning Word Representation Using Box Embeddings [article]

Shib Sankar Dasgupta, Michael Boratko, Shriya Atmakuri, Xiang Lorraine Li, Dhruvesh Patel, Andrew McCallum
2021 arXiv   pre-print
Exciting innovations in the space of representation learning have proposed alternative fundamental representations, such as distributions, hyperbolic vectors, or regions.  ...  Learning vector representations for words is one of the most fundamental topics in NLP, capable of capturing syntactic and semantic relationships useful in a variety of downstream NLP tasks.  ...  Poincaré embeddings (Tifrea et al., 2019) attempt to capture a latent hierarchical graph between words by embedding words as vectors in hyperbolic space.  ... 
arXiv:2106.14361v1 fatcat:lzpcx7qgzrhqxf2plmin6jgray
« Previous Showing results 1 — 15 out of 3,345 results