Similarity-Based Framework for Unsupervised Domain Adaptation: Peer Reviewing Policy for Pseudo-Labeling
-
Published:2023-10-12
Issue:4
Volume:5
Page:1474-1492
-
ISSN:2504-4990
-
Container-title:Machine Learning and Knowledge Extraction
-
language:en
-
Short-container-title:MAKE
Author:
Arweiler Joel1ORCID, Ates Cihan1ORCID, Cerquides Jesus2ORCID, Koch Rainer1, Bauer Hans-Jörg1ORCID
Affiliation:
1. Institute of Thermal Turbomachinery, Karlsruhe Institute of Technology (KIT), 76137 Karlsruhe, Germany 2. Artificial Intelligence Research Institute (IIIA), CSIC, 08193 Bellaterra, Spain
Abstract
The inherent dependency of deep learning models on labeled data is a well-known problem and one of the barriers that slows down the integration of such methods into different fields of applied sciences and engineering, in which experimental and numerical methods can easily generate a colossal amount of unlabeled data. This paper proposes an unsupervised domain adaptation methodology that mimics the peer review process to label new observations in a different domain from the training set. The approach evaluates the validity of a hypothesis using domain knowledge acquired from the training set through a similarity analysis, exploring the projected feature space to examine the class centroid shifts. The methodology is tested on a binary classification problem, where synthetic images of cubes and cylinders in different orientations are generated. The methodology improves the accuracy of the object classifier from 60% to around 90% in the case of a domain shift in physical feature space without human labeling.
Funder
Baden-Württemberg Stiftung
Subject
Artificial Intelligence,Engineering (miscellaneous)
Reference41 articles.
1. Ke, Z., Qiu, D., Li, K., Yan, Q., and Lau, R.W.H. (2020, January 23–28). Guided Collaborative Training for Pixel-Wise Semi-Supervised Learning. Proceedings of the Computer Vision—ECCV 2020, Glasgow, UK. 2. Beyond Supervised Learning: A Computer Vision Perspective;Chum;J. Indian Inst. Sci.,2019 3. A Survey on Semi-, Self- and Unsupervised Learning for Image Classification;Schmarje;IEEE Access,2021 4. Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E., and Garnett, R. (2019). Proceedings of the Advances in Neural Information Processing Systems, Curran Associates, Inc. 5. Zhai, X., Oliver, A., Kolesnikov, A., and Beyer, L. (November, January 27). S4L: Self-Supervised Semi-Supervised Learning. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea.
|
|