Matches in SemOpenAlex for { <https://semopenalex.org/work/W3096168042> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W3096168042 abstract "Efficient modeling of relational data arising in physical, social, and information sciences is challenging due to complicated dependencies within the data. In this work, we build off of semi-implicit graph variational auto-encoders to capture higher-order statistics in a low-dimensional graph latent representation. We incorporate hyperbolic geometry in the latent space through a Poincare embedding to efficiently represent graphs exhibiting hierarchical structure. To address the naive posterior latent distribution assumptions in classical variational inference, we use semi-implicit hierarchical variational Bayes to implicitly capture posteriors of given graph data, which may exhibit heavy tails, multiple modes, skewness, and highly correlated latent structures. We show that the existing semi-implicit variational inference objective provably reduces information in the observed graph. Based on this observation, we estimate and add an additional mutual information term to the semi-implicit variational inference learning objective to capture rich correlations arising between the input and latent spaces. We show that the inclusion of this regularization term in conjunction with the Poincare embedding boosts the quality of learned high-level representations and enables more flexible and faithful graphical modeling. We experimentally demonstrate that our approach outperforms existing graph variational auto-encoders both in Euclidean and in hyperbolic spaces for edge link prediction and node classification." @default.
- W3096168042 created "2020-11-09" @default.
- W3096168042 creator A5010394308 @default.
- W3096168042 creator A5056163027 @default.
- W3096168042 creator A5071614812 @default.
- W3096168042 creator A5081065430 @default.
- W3096168042 creator A5088120102 @default.
- W3096168042 date "2020-10-31" @default.
- W3096168042 modified "2023-09-27" @default.
- W3096168042 title "Hyperbolic Graph Embedding with Enhanced Semi-Implicit Variational Inference" @default.
- W3096168042 cites W1959608418 @default.
- W3096168042 cites W1972673289 @default.
- W3096168042 cites W2018049970 @default.
- W3096168042 cites W2058600246 @default.
- W3096168042 cites W2120340025 @default.
- W3096168042 cites W2125297759 @default.
- W3096168042 cites W2153959628 @default.
- W3096168042 cites W2163922914 @default.
- W3096168042 cites W2783047733 @default.
- W3096168042 cites W2797466362 @default.
- W3096168042 cites W2907140118 @default.
- W3096168042 cites W2962918699 @default.
- W3096168042 cites W2962935371 @default.
- W3096168042 cites W2962936818 @default.
- W3096168042 cites W2963144505 @default.
- W3096168042 cites W2963224980 @default.
- W3096168042 cites W2963757395 @default.
- W3096168042 cites W2963800509 @default.
- W3096168042 cites W2963849380 @default.
- W3096168042 cites W2964282455 @default.
- W3096168042 cites W3081175302 @default.
- W3096168042 cites W3104097132 @default.
- W3096168042 cites W3146803896 @default.
- W3096168042 hasPublicationYear "2020" @default.
- W3096168042 type Work @default.
- W3096168042 sameAs 3096168042 @default.
- W3096168042 citedByCount "0" @default.
- W3096168042 crossrefType "posted-content" @default.
- W3096168042 hasAuthorship W3096168042A5010394308 @default.
- W3096168042 hasAuthorship W3096168042A5056163027 @default.
- W3096168042 hasAuthorship W3096168042A5071614812 @default.
- W3096168042 hasAuthorship W3096168042A5081065430 @default.
- W3096168042 hasAuthorship W3096168042A5088120102 @default.
- W3096168042 hasConcept C11413529 @default.
- W3096168042 hasConcept C132525143 @default.
- W3096168042 hasConcept C154945302 @default.
- W3096168042 hasConcept C2776214188 @default.
- W3096168042 hasConcept C33923547 @default.
- W3096168042 hasConcept C41008148 @default.
- W3096168042 hasConcept C41608201 @default.
- W3096168042 hasConcept C80444323 @default.
- W3096168042 hasConceptScore W3096168042C11413529 @default.
- W3096168042 hasConceptScore W3096168042C132525143 @default.
- W3096168042 hasConceptScore W3096168042C154945302 @default.
- W3096168042 hasConceptScore W3096168042C2776214188 @default.
- W3096168042 hasConceptScore W3096168042C33923547 @default.
- W3096168042 hasConceptScore W3096168042C41008148 @default.
- W3096168042 hasConceptScore W3096168042C41608201 @default.
- W3096168042 hasConceptScore W3096168042C80444323 @default.
- W3096168042 hasLocation W30961680421 @default.
- W3096168042 hasOpenAccess W3096168042 @default.
- W3096168042 hasPrimaryLocation W30961680421 @default.
- W3096168042 hasRelatedWork W1525921156 @default.
- W3096168042 hasRelatedWork W2112104211 @default.
- W3096168042 hasRelatedWork W2150231963 @default.
- W3096168042 hasRelatedWork W2553125747 @default.
- W3096168042 hasRelatedWork W2753380372 @default.
- W3096168042 hasRelatedWork W2948133842 @default.
- W3096168042 hasRelatedWork W2952030117 @default.
- W3096168042 hasRelatedWork W2965198885 @default.
- W3096168042 hasRelatedWork W2970912018 @default.
- W3096168042 hasRelatedWork W3016333069 @default.
- W3096168042 hasRelatedWork W3044177866 @default.
- W3096168042 hasRelatedWork W3044823215 @default.
- W3096168042 hasRelatedWork W3091246499 @default.
- W3096168042 hasRelatedWork W3107396204 @default.
- W3096168042 hasRelatedWork W3124747105 @default.
- W3096168042 hasRelatedWork W3151807233 @default.
- W3096168042 hasRelatedWork W3172566572 @default.
- W3096168042 hasRelatedWork W3191860072 @default.
- W3096168042 hasRelatedWork W3095351248 @default.
- W3096168042 hasRelatedWork W3121098844 @default.
- W3096168042 isParatext "false" @default.
- W3096168042 isRetracted "false" @default.
- W3096168042 magId "3096168042" @default.
- W3096168042 workType "article" @default.