No More Maybe-Arrows: Resolving Causal Uncertainty by Breaking Symmetries

Tingrui Huang, Devendra Singh Dhami · 2026

Read No More Maybe-Arrows: Resolving Causal Uncertainty by Breaking Symmetries by Tingrui Huang, Devendra Singh Dhami (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-01052.txt.

Description

Huang and Dhami resolve causal uncertainty with CausalSAGE, a plan that sharpens partial ancestral graphs into full directed acyclic graphs. Observational data alone pins the true graph only up to a Markov equivalence class, leaving many later uses stuck with maybe-arrows. Their method expands variables into state-level forms, constrains search with structural knowledge and soft priors, and optimizes one smooth objective, combining results while enforcing acyclicity.

Catalog record

CREATORTingrui Huang, Devendra Singh Dhami
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50723
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.01052

Browse the full library · About the Tower of Babel