Metraipolis › archive › dc_376c378dd81e
Graph Domain Adaptation Does Not End with Representation Learning — arXiv:2609.25692
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.25692
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T15:07:31.439Z
- SHA-256 of the archived text
bb7e70f579b35892cc36c501531e912c706b5107e657ea99ee91ea24aae17065- Stable URL of this page
https://metraipolis.com/documento/dc_376c378dd81e
How to cite
Graph Domain Adaptation Does Not End with Representation Learning — arXiv:2609.25692 — arXiv (en), https://arxiv.org/abs/2609.25692 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_376c378dd81e (Metraipolis archive dc_376c378dd81e, sha256 bb7e70f579b35892cc36c501531e912c706b5107e657ea99ee91ea24aae17065)
BibTeX
@misc{dc376c378dd81e,
title = {Graph Domain Adaptation Does Not End with Representation Learning — arXiv:2609.25692},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.25692},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_376c378dd81e; sha256 bb7e70f579b35892cc36c501531e912c706b5107e657ea99ee91ea24aae17065},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25692. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Graph domain adaptation (GDA) transfers knowledge from a labeled source graph to an unlabeled target graph under shifts in both node attributes and graph structure. Existing methods primarily adapt graph representations through propagation redesign, distribution alignment, or source-to-target transition modeling, but still rely on a single graph-propagating path for target prediction. This leaves open whether an adapted graph representation exhausts the predictive evidence available in the target domain, since the graph-aware expert and graph-free local expert may exhibit different failure modes under topological shifts. To address this limitation, we propose EviGDA, an Evidence-Augmented Graph Domain Adaptation framework that complements graph representation adaptation with a graph-free local expert. The graph-aware expert performs message passing and entropy-aware marginal alignment, while the graph-free local expert learns solely from source node features and labels without graph propagation or target alignment. The two experts are optimized independently and combined only at inference through a task-level constant probability mixture, preserving complementary evidence without joint training, learned routing, or target pseudo-labels. Extensive experiments on ten datasets and 16 transfer tasks show that EviGDA outperforms state-of-the-art baselines. Complementary prediction paths add value beyond graph representation alignment. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.25692 — 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 2026-10-08T14:37:29.408Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).