With the increasing acquisition of datasets over time, we now have access to precise and varied descriptions of the world, encompassing a broad range of phenomena. These datasets can be seen as observations from unknown causal generative processes, commonly described by Structural Causal Models (SCMs). Recovering SCMs from observations poses formidable challenges, and often requires us to learn a specific generative model for each dataset. In this work, we propose to learn a \emph{single} model capable of inferring the SCMs in a zero-shot manner. Rather than learning a specific SCM for each dataset, we enable the Fixed-Point Approach (FiP)~\citep{scetbon2024fip} to infer the generative SCMs conditionally on their empirical representations. As a by-product, our approach can perform zero-shot generation of new dataset samples and intervened samples. We demonstrate via experiments that our amortized procedure achieves performances on par with SoTA methods trained specifically for each dataset on both in and out-of-distribution problems. To the best of our knowledge, this is the first time that SCMs are inferred in a zero-shot manner from observations, paving the way for a paradigmatic shift toward the assimilation of causal knowledge across datasets. The code is available on Github.
View on arXiv@article{mahajan2025_2410.06128,
title={ Zero-Shot Learning of Causal Models },
author={ Divyat Mahajan and Jannes Gladrow and Agrin Hilmkil and Cheng Zhang and Meyer Scetbon },
journal={arXiv preprint arXiv:2410.06128},
year={ 2025 }
}