Metraipolis › archive › dc_49fb02174d2f

AutoGrable: What Is a Good Graph for a Table? — arXiv:2608.11431

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.11431
License (as archived): CC0-1.0
Archived: 2026-10-19T05:39:45.698Z
SHA-256 of the archived text: b64ce3c53f510962f337e76d5db42698ae460606d8bb59e8919b89f29cf5c6f1
Stable URL of this page: https://metraipolis.com/documento/dc_49fb02174d2f

How to cite

AutoGrable: What Is a Good Graph for a Table? — arXiv:2608.11431 — arXiv (en), https://arxiv.org/abs/2608.11431 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_49fb02174d2f (Metraipolis archive dc_49fb02174d2f, sha256 b64ce3c53f510962f337e76d5db42698ae460606d8bb59e8919b89f29cf5c6f1)

BibTeX

@misc{dc49fb02174d2f,
  title = {AutoGrable: What Is a Good Graph for a Table? — arXiv:2608.11431},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.11431},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_49fb02174d2f; sha256 b64ce3c53f510962f337e76d5db42698ae460606d8bb59e8919b89f29cf5c6f1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11431. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Graph learning presupposes a graph, and tables and relational databases do not come with one. Applying a GNN to them requires deciding which entities become nodes, which of them to connect, and through which relations---a decision made by hand, by schema heuristics, or by training a model on every candidate graph and keeping the best. We give a criterion that requires no trained graph model. In the minimal table-to-graph abstraction each row is a node, so a message-passing GNN, bounded by 1-WL, sees a construction only as a partition of the rows into colour-refinement classes: a construction is good for a task when that partition separates rows with different labels and does not split rows that share one. AutoGrable turns this criterion into a construction procedure. For incidence constructions the partition is fixed by the selected columns, so building a graph reduces to choosing them, and we score a candidate subset by a label-alignment risk: the held-out risk of the best predictor constant on its blocks, penalised by an occupancy term measuring how thinly the blocks are populated. The score materialises no graph and trains no GNN, so AutoGrable can search the space of subsets greedily and cheaply, and returns the resulting grable for single tables and for foreign-key schemas alike. Our experiments show that over a space of candidate graphs the score discards a large fraction while retaining the best; that AutoGrable recovers the columns that generate the label on controlled tasks and outperforms fixed, random, and task-aware constructors on real tasks under a fixed predictor; and that it is the only method compared that can decline to build a graph when none helps.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.11431 — 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 2026-10-19T05:18:32.464Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).