DyVGRNN: DYnamic mixture variational graph recurrent neural networks
Although graph representation learning has been studied extensively in static graph settings, dynamic graphs are less investigated in this context. This paper proposes a novel integrated variational framework called DYnamic mixture Variational Graph Recurrent Neural Networks (DyVGRNN), which consist...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Elsevier
2023
|
| _version_ | 1797110260225277952 |
|---|---|
| author | Niknam, G Molaei, S Zare, H Pan, S Jalili, M Zhu, T Clifton, D |
| author_facet | Niknam, G Molaei, S Zare, H Pan, S Jalili, M Zhu, T Clifton, D |
| author_sort | Niknam, G |
| collection | OXFORD |
| description | Although graph representation learning has been studied extensively in static graph settings, dynamic graphs are less investigated in this context. This paper proposes a novel integrated variational framework called DYnamic mixture Variational Graph Recurrent Neural Networks (DyVGRNN), which consists of extra latent random variables in structural and temporal modelling. Our proposed framework comprises an integration of Variational Graph Auto-Encoder (VGAE) and Graph Recurrent Neural Network (GRNN) by exploiting a novel attention mechanism. The Gaussian Mixture Model (GMM) and the VGAE framework are combined in DyVGRNN to model the multimodal nature of data, which enhances performance. To consider the significance of time steps, our proposed method incorporates an attention-based module. The experimental results demonstrate that our method greatly outperforms state-of-the-art dynamic graph representation learning methods in terms of link prediction and clustering. |
| first_indexed | 2024-03-07T07:52:30Z |
| format | Journal article |
| id | oxford-uuid:1eac2ae9-51a1-4da8-ab2f-6e0bfd92b4d9 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:52:30Z |
| publishDate | 2023 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:1eac2ae9-51a1-4da8-ab2f-6e0bfd92b4d92023-08-01T12:46:26ZDyVGRNN: DYnamic mixture variational graph recurrent neural networksJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:1eac2ae9-51a1-4da8-ab2f-6e0bfd92b4d9EnglishSymplectic ElementsElsevier2023Niknam, GMolaei, SZare, HPan, SJalili, MZhu, TClifton, DAlthough graph representation learning has been studied extensively in static graph settings, dynamic graphs are less investigated in this context. This paper proposes a novel integrated variational framework called DYnamic mixture Variational Graph Recurrent Neural Networks (DyVGRNN), which consists of extra latent random variables in structural and temporal modelling. Our proposed framework comprises an integration of Variational Graph Auto-Encoder (VGAE) and Graph Recurrent Neural Network (GRNN) by exploiting a novel attention mechanism. The Gaussian Mixture Model (GMM) and the VGAE framework are combined in DyVGRNN to model the multimodal nature of data, which enhances performance. To consider the significance of time steps, our proposed method incorporates an attention-based module. The experimental results demonstrate that our method greatly outperforms state-of-the-art dynamic graph representation learning methods in terms of link prediction and clustering. |
| spellingShingle | Niknam, G Molaei, S Zare, H Pan, S Jalili, M Zhu, T Clifton, D DyVGRNN: DYnamic mixture variational graph recurrent neural networks |
| title | DyVGRNN: DYnamic mixture variational graph recurrent neural networks |
| title_full | DyVGRNN: DYnamic mixture variational graph recurrent neural networks |
| title_fullStr | DyVGRNN: DYnamic mixture variational graph recurrent neural networks |
| title_full_unstemmed | DyVGRNN: DYnamic mixture variational graph recurrent neural networks |
| title_short | DyVGRNN: DYnamic mixture variational graph recurrent neural networks |
| title_sort | dyvgrnn dynamic mixture variational graph recurrent neural networks |
| work_keys_str_mv | AT niknamg dyvgrnndynamicmixturevariationalgraphrecurrentneuralnetworks AT molaeis dyvgrnndynamicmixturevariationalgraphrecurrentneuralnetworks AT zareh dyvgrnndynamicmixturevariationalgraphrecurrentneuralnetworks AT pans dyvgrnndynamicmixturevariationalgraphrecurrentneuralnetworks AT jalilim dyvgrnndynamicmixturevariationalgraphrecurrentneuralnetworks AT zhut dyvgrnndynamicmixturevariationalgraphrecurrentneuralnetworks AT cliftond dyvgrnndynamicmixturevariationalgraphrecurrentneuralnetworks |