Metraipolis › archive › dc_919fbcd59edf

DeDiCa: A Scalable Deconfounding Discrete Causal Generative Model — arXiv:2610.23272

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.23272
License (as archived): CC0-1.0
Archived: 2026-10-19T08:48:42.024Z
SHA-256 of the archived text: f7b6287d1d6c4d1be107e591c7c7fb70c27248ab3289e49bd607c1dde634e522
Stable URL of this page: https://metraipolis.com/documento/dc_919fbcd59edf

How to cite

DeDiCa: A Scalable Deconfounding Discrete Causal Generative Model — arXiv:2610.23272 — arXiv (en), https://arxiv.org/abs/2610.23272 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_919fbcd59edf (Metraipolis archive dc_919fbcd59edf, sha256 f7b6287d1d6c4d1be107e591c7c7fb70c27248ab3289e49bd607c1dde634e522)

BibTeX

@misc{dc919fbcd59edf,
  title = {DeDiCa: A Scalable Deconfounding Discrete Causal Generative Model — arXiv:2610.23272},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23272},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_919fbcd59edf; sha256 f7b6287d1d6c4d1be107e591c7c7fb70c27248ab3289e49bd607c1dde634e522},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23272. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We introduce DeDiCa, a practical framework for efficient causal inference with discrete variables and hidden confounding that revisits classical probabilistic modeling through a tensorized parameterization of causal Bayesian networks, enabling tractable inference. Given a causal graph, DeDiCa is trained once on observational data by maximum likelihood, exactly marginalizing over hidden confounders, and can then estimate any (proxy-)identifiable interventional query. Despite this deliberately simple parameterization, DeDiCa matches state-of-the-art neural causal generative models in ATE estimation, while supporting memory-efficient CPU training and exact interventional inference in milliseconds. Compared with causal foundation models, DeDiCa scales more effectively with additional observational samples and achieves competitive accuracy at substantially lower inference time and memory. DeDiCa provides a tractable, accurate, and scalable approach to causal inference in discrete settings with hidden confounding.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.23272 — 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 2026-10-19T08:42:40.270Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).