Bivariate Causal Discovery Using Rate-Distortion MDL: An Information Dimension Approach

Tiago Brogueira, Mário A. T. Figueiredo · 2026

Read Bivariate Causal Discovery Using Rate-Distortion MDL: An Information Dimension Approach by Tiago Brogueira, Mário A. T. Figueiredo (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-2604-05829.txt.

Description

Brogueira and Figueiredo sharpen two-variable causal discovery by pricing the cause itself. Usual minimum-description tests judge only the mechanism, leaving the cause's own cost out. Their rate-distortion MDL prices the cause by the smallest rate for a typical distortion, set via histogram density rules and information dimension. Tests on the Tübingen set prove competitive.

Catalog record

CREATORTiago Brogueira, Mário A. T. Figueiredo
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55162
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.05829

Browse the full library · About the Tower of Babel