Classification Certainty Maximization for Unsupervised Domain Adaptation
- 6 February 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 33 (8) , 4232-4243
Abstract
The bi-classifier paradigm is a common practice in unsupervised domain adaptation (UDA), where two classifiers are leveraged to guide the model to learn domain invariant features. Previous approaches only focused on the consistency of the outputs between classifiers, but ignored the classification certainty of each classifier. Therefore, existing methods in some cases may mislead the classifiers into the wrong direction of ambiguous outputs and subsequently undermine the discriminability. To challenge this problem, in this paper we propose Classification Certainty Maximization (CCM) which considers both the joint certainty between classifiers and the individual certainty of each classifier via a novel formulation, and derived their optimal weight ratios by theoretical derivation from the perspective of gradient. In addition, we also propose a dynamic centroid update strategy to mitigate the domain gaps at the feature level. Extensive experiments on four widely used UDA datasets show that CCM performs better than the existing state-of-the-art domain adaptation methods. Notably, our dynamic centroid update strategy can be used as a plug-and-play module for existing bi-classifier domain adaptation methods to boost classification accuracy.Keywords
Funding Information
- National Natural Science Foundation of China (62176042, 62073059)
- China Computer Federation (CCF)-Baidu Open Fund (2021PP15002000)
- Basic and Applied Basic Research Foundation of Guangdong Province (2021B1515140013)
- Fundamental Research Funds for the Central Universities (ZYGX2021YGCX016)
This publication has 40 references indexed in Scilit:
- Joint Adversarial Domain Adaptation2019
- Cycle-consistent Conditional Adversarial Transfer Networks2019
- Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-training2018
- Maximum Classifier Discrepancy for Unsupervised Domain Adaptation2018
- Learning to Adapt Structured Output Space for Semantic Segmentation2018
- Deep Residual Learning for Image Recognition2016
- Structured Domain AdaptationIEEE Transactions on Circuits and Systems for Video Technology, 2016
- Hybrid CNN and Dictionary-Based Models for Scene Recognition and Domain AdaptationIEEE Transactions on Circuits and Systems for Video Technology, 2015
- Large-Scale Machine Learning with Stochastic Gradient Descent2010
- A theory of learning from different domainsMachine Learning, 2009