Causal‐Invariance Transfer Learning via Process‐Topology‐Aware Invertible Flows for Weak Cold‐Start Quality Prediction in Multicondition Process Manufacturing
Advanced Intelligent Systems · 17.11.2026
DOI: 10.1002/aisy.70631 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advanced Intelligent Systems, online erschienen am 17.11.2026. Verfasst von Yiquan An und 5 weiteren Autor:innen. Die Arbeit nennt 89 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Major Project.
Abstract (Originalsprache)
Weak cold‐start quality prediction in process industries faces two coupled challenges: new grades provide only a handful of labels, and cross‐condition shifts make observed correlations environment‐dependent. The task is formulated as preserving transferable quality‐formation mechanisms while suppressing environment‐specific disturbances, and an end‐to‐end framework is proposed that uses mechanism stability as the transfer criterion. In an invertible representation space, structured routing disentangles stable predictive information from spurious variation under a process‐chain topology prior. Multienvironment conditional likelihoods model admissible environment‐dependent variation in the base distribution, and adversarial regularization suppresses residual environment‐discriminative information. Under weak supervision, causally consistent augmentation expands target‐domain support, and likelihood‐based screening limits off‐manifold drift and extrapolation. Conditional quantile modeling captures central tendency and tail uncertainty, yielding calibrated prediction intervals. Evaluation uses a hot‐rolling dataset of 3391 coils and 129 variables spanning steelmaking to coiling, covering 17 grades, seven of them target grades with fewer than 50 labels each. Against thirteen baselines, the framework achieves the lowest mean absolute percentage errors, 1.2%–4.3% for tensile strength, yield strength, and elongation, a 37%–45% average reduction. The nominal 80% prediction interval attains 82% empirical coverage. The framework therefore supports quality assessment and risk screening for new grades before inspection data accumulate.
Förderung
- National Science and Technology Major Project
Zitieren
APA
An, Y., Smedt, J. D., Lv, Z., Li, Y., Sun, X., & Bérard, R. (2026). Causal‐Invariance Transfer Learning via Process‐Topology‐Aware Invertible Flows for Weak Cold‐Start Quality Prediction in Multicondition Process Manufacturing. Advanced Intelligent Systems. https://doi.org/10.1002/aisy.70631
Vancouver
An Y, Smedt JD, Lv Z, Li Y, Sun X, Bérard R. Causal‐Invariance Transfer Learning via Process‐Topology‐Aware Invertible Flows for Weak Cold‐Start Quality Prediction in Multicondition Process Manufacturing. Advanced Intelligent Systems. 2026. doi:10.1002/aisy.70631
BibTeX
@article{An2026,
author = {An, Yiquan and Smedt, Johannes De and Lv, Zhimin and Li, Yihang and Sun, Xi and Bérard, Romain},
title = {Causal‐Invariance Transfer Learning via Process‐Topology‐Aware Invertible Flows for Weak Cold‐Start Quality Prediction in Multicondition Process Manufacturing},
journal = {Advanced Intelligent Systems},
year = {2026},
doi = {10.1002/aisy.70631}
}
EUVIMED