Multimodal deep learning on hypergraphs
| Authors | |
|---|---|
| Supervisors | |
| Award date | 17-06-2022 |
| Number of pages | 114 |
| Organisations |
|
| Abstract | This thesis investigates the potential of hypergraphs for capturing higher-order relations between objects in a multimodal dataset. These relations are often sub-optimally represented by pairwise connections used in a graph. Hence, in order to unlock the full potential of relational information within a multimodal dataset, this thesis proposes several geometric deep learning approaches for capturing and learning higher-order relations. |
| Document type | PhD thesis |
| Language | English |
| Downloads | |
| Permalink to this page | |
