PACER: Acyclic Causal Discovery from Large-Scale Interventional Data
The paper introduces PACER, a new framework for causal discovery that ensures acyclicity in directed acyclic graphs. It addresses limitations of existing methods by allowing direct optimization over valid causal structures. Empirical results show that PACER outperforms state-of-the-art methods while being scalable to large networks.
- ▪PACER stands for Perturbation-driven Acyclic Causal Edge Recovery.
- ▪The framework allows for a unified treatment of observational and interventional data.
- ▪PACER achieves significant computational gains, especially for linear-Gaussian mechanisms.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Machine Learning arXiv:2605.15353 (cs) [Submitted on 14 May 2026] Title:PACER: Acyclic Causal Discovery from Large-Scale Interventional Data Authors:Ramon Viñas Torné, Sílvia Fàbregas Salazar, Soyon Park, Ivo Alexander Ban, Artyom Gadetsky, Nikita Doikov, Maria Brbić View a PDF of the paper titled PACER: Acyclic Causal Discovery from Large-Scale Interventional Data, by Ramon Vi\~nas Torn\'e and 6 other authors View PDF HTML (experimental) Abstract:Inferring the structure of directed acyclic graphs (DAGs) from data is a central challenge in causal discovery, particularly in modern high-dimensional settings where large-scale interventional data are increasingly available.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
