Journal
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
Volume 33, Issue 8, Pages 3857-3871Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TNNLS.2021.3054665
Keywords
Transfer learning; Knowledge transfer; Adaptation models; Data models; Correlation; Technological innovation; Learning systems; Concept drift; ensemble learning; negative transfer; positive transfer; transfer learning
Categories
Funding
- National Natural Science Foundation of China [61672444, 61988101]
- Hong Kong Baptist University (HKBU) [RCFNRA-IG/18-19/SCI/03, RC-IRCMs/18-19/SCI/01]
- Innovation and Technology Fund of Innovation and Technology Commission of the Government of the Hong Kong SAR [ITS/339/18]
- National Key Research and Development Program of China [2018YFB1701104]
- Xingliao Plan of Liaoning Province [XLYC1808001]
- Science and Technology Program of Liaoning Province [2020JH2/10500001, 2020JH1/10100008]
- Research Grants Council of the Hong Kong SAR [CityU11202418, CityU11209219]
Ask authors/readers for more resources
This article proposes a hybrid ensemble approach to address the concept drift-tolerant transfer learning problem, adapting target domain models to new environments through class-wise weighted ensemble. The approach assigns weight vectors for classifiers from previous data chunks, allowing each class of current data to leverage historical knowledge independently.
Existing transfer learning methods that focus on problems in stationary environments are not usually applicable to dynamic environments, where concept drift may occur. To the best of our knowledge, the concept drift-tolerant transfer learning (CDTL), whose major challenge is the need to adapt the target model and knowledge of source domains to the changing environments, has yet to be well explored in the literature. This article, therefore, proposes a hybrid ensemble approach to deal with the CDTL problem provided that data in the target domain are generated in a streaming chunk-by-chunk manner from nonstationary environments. At each time step, a class-wise weighted ensemble is presented to adapt the model of target domains to new environments. It assigns a weight vector for each classifier generated from the previous data chunks to allow each class of the current data leveraging historical knowledge independently. Then, a domain-wise weighted ensemble is introduced to combine the source and target models to select useful knowledge of each domain. The source models are updated with the source instances performed by the proposed adaptive weighted CORrelation ALignment (AW-CORAL). AW-CORAL iteratively minimizes domain discrepancy meanwhile decreases the effect of unrelated source instances. In this way, positive knowledge of source domains can be potentially promoted while negative knowledge is reduced. Empirical studies on synthetic and real benchmark data sets demonstrate the effectiveness of the proposed algorithm.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available