One-shot multi-label causal discovery in high-dimensional event sequences

  • Understanding causality in event sequences with thousands of sparse event types is critical in domains such as healthcare, cybersecurity or vehicle diagnostics, yet current methods fail to scale. We present OSCAR, a one-shot causal autoregressive method that infers per-sequence Markov Boundaries using two pretrained Transformers as density estimators. This enables efficient, parallel causal discovery without costly global CI testing. On a real-world automotive dataset with 29,100 events and 474 labels, OSCAR recovers interpretable causal structures in minutes, while classical methods fail to scale— enabling practical scientific diagnostics at production scale.

Download full text files

Export metadata

Statistics

Number of document requests

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Hugo MathGND, Robin SchönORCiDGND, Rainer LienhartORCiDGND
URN:urn:nbn:de:bvb:384-opus4-1262738
Frontdoor URLhttps://opus.bibliothek.uni-augsburg.de/opus4/126273
URL:https://openreview.net/forum?id=z7NT8vGWC2
Parent Title (English):NeurIPS 2025 Workshop on CauScien: Uncovering Causality in Science, 6 December 2025, San Diego, CA, USA
Type:Conference Proceeding
Language:English
Date of Publication (online):2025/11/11
Year of first Publication:2025
Publishing Institution:Universität Augsburg
Release Date:2025/11/12
Institutes:Fakultät für Angewandte Informatik
Fakultät für Angewandte Informatik / Institut für Informatik
Fakultät für Angewandte Informatik / Institut für Informatik / Lehrstuhl für Maschinelles Lernen und Maschinelles Sehen
Dewey Decimal Classification:0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 004 Datenverarbeitung; Informatik
Licence (German):CC-BY 4.0: Creative Commons: Namensnennung

OPUS4 Logo