Metraipolis › archive › dc_f1745847db98
Equivariant Neural Primal-Dual Assignment for Maximum Common Edge Subgraphs — arXiv:2609.32661
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32661
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:36:45.167Z
- SHA-256 of the archived text
7aa525a502e7868458cbb0c3d21af17fdde30a2c6f847024ae3ad5f28917b60f- Stable URL of this page
https://metraipolis.com/documento/dc_f1745847db98
How to cite
Equivariant Neural Primal-Dual Assignment for Maximum Common Edge Subgraphs — arXiv:2609.32661 — arXiv (en), https://arxiv.org/abs/2609.32661 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f1745847db98 (Metraipolis archive dc_f1745847db98, sha256 7aa525a502e7868458cbb0c3d21af17fdde30a2c6f847024ae3ad5f28917b60f)
BibTeX
@misc{dcf1745847db98,
title = {Equivariant Neural Primal-Dual Assignment for Maximum Common Edge Subgraphs — arXiv:2609.32661},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32661},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f1745847db98; sha256 7aa525a502e7868458cbb0c3d21af17fdde30a2c6f847024ae3ad5f28917b60f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32661. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Maximum common edge subgraph (MCES) matching finds a partial vertex correspondence between two labeled graphs that preserves as many labeled edges as possible. Molecular similarity search requires matching many graph pairs, making the cost of repeated queries important. The strongest baseline attains accurate MCES solutions but trains a separate network for each pair. We introduce Equivariant Neural Primal-Dual Assignment (ENPDA), which learns a shared matching policy and applies it to new pairs without further training, answering queries roughly three orders of magnitude faster and recovering its training cost after a few dozen queries. The policy recomputes exact objective marginals for candidate matches and learns corrections and step sizes that update their scores. Target prices respond to competition when several source vertices favor the same target. Four update rounds and a Hungarian projection produce a partial one-to-one matching. We prove per-pair guarantees that hold for any network parameters. In exact arithmetic, reordering either graph permutes the assignment and price states, the projected matching is one-to-one, and repaired prices give a valid MCES upper bound. Subtracting the preserved-edge count bounds the optimality gap; combined with structural caps, these certificates prove global optimality for 60 of 291 native test pairs. On three molecular benchmarks with disjoint train/validation/test splits, ENPDA improves over an analytic counterpart with the same update and projection budget by 7.4-8.6 accuracy points; after one second of refinement search, 2.5-3 points of the gain remain. Transferred without fine-tuning to edge-deletion tasks from social and protein graphs, the policy gains 9.1-17.6 points over the analytic counterpart. When output matchings must keep aromatic rings intact, ENPDA recovers more reference bonds than the baselines on all three datasets. 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/2609.32661 — 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 2026-10-07T22:14:27.737Z 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).