Metraipolis › archive › dc_80d12e2648ea

When Reconstruction Is Not Enough: Doubly Robust Counterfactual and Interventional Mediation Analysis with Learned Multimodal Representations — arXiv:2610.30220

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.30220
License (as archived): CC0-1.0
Archived: 2026-10-25T20:57:07.162Z
SHA-256 of the archived text: 83f4bfa712bc05b328cf5fe22770efb5f879d9acaa58dfa576e1a84acee80f65
Stable URL of this page: https://metraipolis.com/documento/dc_80d12e2648ea

How to cite

When Reconstruction Is Not Enough: Doubly Robust Counterfactual and Interventional Mediation Analysis with Learned Multimodal Representations — arXiv:2610.30220 — arXiv (en), https://arxiv.org/abs/2610.30220 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_80d12e2648ea (Metraipolis archive dc_80d12e2648ea, sha256 83f4bfa712bc05b328cf5fe22770efb5f879d9acaa58dfa576e1a84acee80f65)

BibTeX

@misc{dc80d12e2648ea,
  title = {When Reconstruction Is Not Enough: Doubly Robust Counterfactual and Interventional Mediation Analysis with Learned Multimodal Representations — arXiv:2610.30220},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30220},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_80d12e2648ea; sha256 83f4bfa712bc05b328cf5fe22770efb5f879d9acaa58dfa576e1a84acee80f65},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30220. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learned representations of high-dimensional multimodal health records are increasingly used to adjust for confounding, yet they are usually trained to reconstruct the data rather than preserve the information that causal identification requires. We study counterfactual outcomes under alternative treatments, including phenotype-specific effects and patient-level counterfactual risk profiles of the kind digital twins provide, and interventional direct and indirect effects under exposure-induced intermediate confounding. We give conditions under which a representation preserves identification and show that interventional mediation is more demanding than counterfactual comparison. We prove that no covariate-only training objective, whether reconstruction, contrastive, or physics-informed, can certify that a representation is sufficient, and that, in a linear structural model, a representation can lose arbitrarily little reconstruction accuracy yet induce arbitrarily large bias. We decompose the error of cross-fitted doubly robust estimators into a first-order representation bias, which neither double robustness nor cross-fitting removes, and a second-order remainder of nuisance-estimation errors, so coverage worsens as the sample size grows when the representation is insufficient. We propose a workflow combining modality-specific variational autoencoders trained outside the estimation sample, cross-fitted doubly robust estimators, locked phenotypes, select-then-confirm pathway analysis, portable representations, and diagnostics for representation insufficiency, exposure leakage, and sensitivity. In simulations, discarding a rare, moderately strong confounder gave indirect-effect intervals with 2% coverage at n = 3,000; reconstruction loss missed the problem, and a simple predictive diagnostic flagged it in every replicate.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Methodology (stat.ME); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.30220 — 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 2026-10-25T20:47:56.972Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).