Fuzzy Multiple-Source Transfer Learning
- 11 November 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Fuzzy Systems
- Vol. 28 (12) , 3418-3431
Abstract
Transfer learning is gaining increasing attention due to its ability to leverage previously acquired knowledge to assist in completing a prediction task in a related domain. Fuzzy transfer learning, which is based on fuzzy systems and particularly fuzzy rule-based models, was developed due to its capacity to deal with uncertainty. However, one issue with fuzzy transfer learning, even in the area of general transfer learning, has not been resolved: how to combine and then use knowledge when multiple source domains are available. This study presents new methods for merging fuzzy rules from multiple domains for regression tasks. Two different settings are separately explored: homogeneous and heterogeneous space. In homogeneous situations, knowledge from the source domains is merged in the form of fuzzy rules. In heterogeneous situations, knowledge is merged in the form of both data and fuzzy rules. Experiments on both synthetic and real-world datasets provide insights into the scope of applications suitable for the proposed methods and validate their effectiveness through comparisons with other state-of-the-art transfer learning methods. An analysis of parameter sensitivity is also included.Keywords
Funding Information
- Australian Research Council (170101632)
This publication has 34 references indexed in Scilit:
- A survey of transfer learning for collaborative recommendation with auxiliary dataNeurocomputing, 2016
- Transfer learning using computational intelligence: A surveyKnowledge-Based Systems, 2015
- Fuzzy Transfer Learning: Methodology and applicationInformation Sciences, 2015
- Multistep Fuzzy Bridged Refinement Domain Adaptation Algorithm and Its Application to Bank Failure PredictionIEEE Transactions on Fuzzy Systems, 2015
- Transfer Learning for Visual Categorization: A SurveyIEEE Transactions on Neural Networks and Learning Systems, 2014
- Transfer Learning with Multiple Sources via Consensus Regularized Autoencoders2014
- Transfer learning for activity recognition: a surveyKnowledge and Information Systems, 2013
- Multi-Transfer: Transfer Learning with Multiple Views and Multiple Sources2013
- FUZZY BRIDGED REFINEMENT DOMAIN ADAPTATION: LONG-TERM BANK FAILURE PREDICTIONInternational Journal of Computational Intelligence and Applications, 2013
- Takagi-Sugeno fuzzy modeling incorporating input variables selectionIEEE Transactions on Fuzzy Systems, 2002