One-Shot Multi-Label Causal Discovery in High-Dimensional Event Sequences

Hugo Math, Robin Schön, Rainer Lienhart

Keywords: causal discovery, event sequences, multi-label causal discovery, density estimation, time-series, structure learning

TL;DR: We introduce OSCAR, a one-shot causal discovery method that identifies interpretable event-to-label causal structure in high-dimensional event sequences using pretrained Transformers, scaling to thousands of nodes.

Abstract: 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.

Submission Number: 6