Metraipolis › archive › dc_369b36bc85cf
Gromov-Wasserstein Distillation for Inductive Multi-View Embedding — arXiv:2609.40047
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.40047
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T13:46:51.269Z
- SHA-256 of the archived text
dd8633a4fa6c3e65ce2f924ecd8a551f6759ac08bcb04ac9f6f737548e5d5449- Stable URL of this page
https://metraipolis.com/documento/dc_369b36bc85cf
How to cite
Gromov-Wasserstein Distillation for Inductive Multi-View Embedding — arXiv:2609.40047 — arXiv (en), https://arxiv.org/abs/2609.40047 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_369b36bc85cf (Metraipolis archive dc_369b36bc85cf, sha256 dd8633a4fa6c3e65ce2f924ecd8a551f6759ac08bcb04ac9f6f737548e5d5449)
BibTeX
@misc{dc369b36bc85cf,
title = {Gromov-Wasserstein Distillation for Inductive Multi-View Embedding — arXiv:2609.40047},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.40047},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_369b36bc85cf; sha256 dd8633a4fa6c3e65ce2f924ecd8a551f6759ac08bcb04ac9f6f737548e5d5449},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40047. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Gromov-Wasserstein multidimensional scaling (GW-MDS) learns low-dimensional representations from relational data but remains transductive, providing no explicit mapping for unseen samples. We introduce an inductive framework based on barycentric distillation. A GW-MDS teacher learns a latent support and an optimal transport plan from the training data, and barycentric projection converts the resulting coupling into sample-aligned targets. A neural student then learns an explicit out-of-sample mapping, avoiding additional relational-matrix construction and GW optimization at inference. We formulate the approach for single-view data and extend it to Mean-GWMDS and Multi-GWMDS teachers through consensus and selected-projection targets learned by a multi-view student with view-specific encoders. We also investigate a direct neural baseline trained solely with a GW objective. Experiments on synthetic and real-world data using Euclidean, geodesic, and cosine relations show that the distilled models preserve the teacher geometry on unseen samples and consistently outperform direct neural GW training in sample-indexed relational preservation. These results establish barycentric projection as an effective bridge between transductive GW embeddings and inductive neural mappings. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG). Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML). Pagina del lavoro: https://arxiv.org/abs/2609.40047 — 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 2026-10-07T13:18:10.499Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).