Universität zu Köln

Li, Ziyue ORCID: 0000-0003-4983-9352, Yan, Hao ORCID: 0000-0002-4322-7323, Tsung, Fugee ORCID: 0000-0002-0575-8254 and Zhang, Ke ORCID: 0000-0002-7827-1770 (2022). Profile Decomposition Based Hybrid Transfer Learning for Cold-Start Data Anomaly Detection. ACM Trans. Knowl. Discov. Data, 16 (6). NEW YORK: ASSOC COMPUTING MACHINERY. ISSN 1556-472X

Full text not available from this repository.

Abstract

Anomaly detection is an essential task for quality management in smart manufacturing. An accurate data-driven detection method usually needs enough data and labels. However, in practice, there commonly exist newly set-up processes in manufacturing, and they only have quite limited data available for analysis. Borrowing the name from the recommender system, we call this process a cold-start process. The sparsity of anomaly, the deviation of the profile, and noise aggravate the detection difficulty. Transfer learning could help to detect anomalies for cold-start processes by transferring the knowledge from more experienced processes to the new processes. However, the existing transfer learning and multi-task learning frameworks are established on task- or domain-level relatedness. We observe instead, within a domain, some components (background and anomaly) share more commonality, others (profile deviation and noise) not. To this end, we propose a more delicate component-level transfer learning scheme, i.e., decomposition-based hybrid transfer learning (DHTL): It first decomposes a domain (e.g., a data source containing profiles) into different components (smooth background, profile deviation, anomaly, and noise); then, each component's transferability is analyzed by expert knowledge; Lastly, different transfer learning techniques could be tailored accordingly. We adopted the Bayesian probabilistic hierarchical model to formulate parameter transfer for the background, and L2,1 + L1-norm to formulate low dimension feature-representation transfer for the anomaly. An efficient algorithm based on Block Coordinate Descend is proposed to learn the parameters. A case study based on glass coating pressure profiles demonstrates the improved accuracy and completeness of detected anomaly, and a simulation demonstrates the fidelity of the decomposition results.

Item Type:	Journal Article
Creators:	Creators Email ORCID ORCID Put Code Li, Ziyue UNSPECIFIED orcid.org/0000-0003-4983-9352 UNSPECIFIED Yan, Hao UNSPECIFIED orcid.org/0000-0002-4322-7323 UNSPECIFIED Tsung, Fugee UNSPECIFIED orcid.org/0000-0002-0575-8254 UNSPECIFIED Zhang, Ke UNSPECIFIED orcid.org/0000-0002-7827-1770 UNSPECIFIED
URN:	urn:nbn:de:hbz:38-676834
DOI:	10.1145/3530990
Journal or Publication Title:	ACM Trans. Knowl. Discov. Data
Volume:	16
Number:	6
Date:	2022
Publisher:	ASSOC COMPUTING MACHINERY
Place of Publication:	NEW YORK
ISSN:	1556-472X
Language:	English
Faculty:	Unspecified
Divisions:	Unspecified
Subjects:	no entry
Uncontrolled Keywords:	Keywords Language SPARSE; MIXTURE Multiple languages Computer Science, Information Systems; Computer Science, Software Engineering Multiple languages
URI:	http://kups.ub.uni-koeln.de/id/eprint/67683