A New Approach to Backtracking Counterfactual Explanations: A Unified Causal Framework for Efficient Model Interpretability

Pouria Fatemi, Ehsan Sharifian, Mohammad Hossein Yassaee · 2025

Read A New Approach to Backtracking Counterfactual Explanations: A Unified Causal Framework for Efficient Model Interpretability by Pouria Fatemi, Ehsan Sharifian, Mohammad Hossein Yassaee (2025) 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-2505-02435.txt.

Description

Pouria Fatemi, Ehsan Sharifian, and Mohammad Hossein Yassaee present a 2025 method for explaining model calls with what-if cases that respect cause and effect. Plain what-if tools swap inputs freely and can craft cases that could never occur. Causal versions fix that flaw but cost much compute. The gap calls for a fast, causal route.

Their method, called BRACE, works backward through a causal model to build the what-if case. It claims deeper insight into why a model ruled as it did. The authors show their frame covers older methods as special cases. Tests back the speed and depth claims. The paper also lays out where older methods fall short, to frame the gain.

Catalog record

CREATORPouria Fatemi, Ehsan Sharifian, Mohammad Hossein Yassaee
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-02435

Browse the full library · About the Tower of Babel