Metraipolis › archive › dc_524440489e73
Temporal Heterogeneous Graph Pretraining for Relational Deep Learning — arXiv:2609.35219
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35219
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:38:56.638Z
- SHA-256 of the archived text
bb43d1db7d9672fef08bb18d5feb07c6b91b4995b41d2a70f740583826c3c62c- Stable URL of this page
https://metraipolis.com/documento/dc_524440489e73
How to cite
Temporal Heterogeneous Graph Pretraining for Relational Deep Learning — arXiv:2609.35219 — arXiv (en), https://arxiv.org/abs/2609.35219 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_524440489e73 (Metraipolis archive dc_524440489e73, sha256 bb43d1db7d9672fef08bb18d5feb07c6b91b4995b41d2a70f740583826c3c62c)
BibTeX
@misc{dc524440489e73,
title = {Temporal Heterogeneous Graph Pretraining for Relational Deep Learning — arXiv:2609.35219},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35219},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_524440489e73; sha256 bb43d1db7d9672fef08bb18d5feb07c6b91b4995b41d2a70f740583826c3c62c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35219. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Relational deep learning models database rows and foreign-key links as a heterogeneous graph for prediction from record attributes and relational context. These graphs contain two distinct temporal signals: record age changes with the prediction cutoff, while intervals between observed records remain fixed. Prior work often treats time as a single signal or studies temporal representation and pretraining separately. We investigate how explicitly encoding both signals affects temporal pretraining for downstream tasks. Our framework combines Multi-scale Time Encoding, which captures record age using learnable time scales and type-specific projections, with Rotary Time Encoding, which represents signed inter-record intervals through rotary transformations during graph propagation. We pair these encodings with three self-supervised objectives: historical relation recovery, horizon-aware future relation activity prediction, and temporal subgraph contrast. All inputs respect their observation cutoffs. Pretraining proceeds in two stages: subgraph contrast first learns neighborhood representations, followed by refinement through either relation recovery or future activity prediction. We evaluate on five RelBench datasets across 11 classification and regression tasks using heterogeneous GNN and graph Transformer backbones. With both encodings, the best evaluated staged schedules improve over supervised training with the same encodings by 3.02% and 1.06% on the two backbones, respectively, and over controls without pretraining or either encoding by 3.24% and 2.37%. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.35219 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T22:14:28.385Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).