The 8th DeSI Lab Seminar
The 8th DeSI Lab Seminar on Fuzzy rule-based domain adaptation in homogeneous and heterogeneous spaces.
Title: Fuzzy rule-based domain adaptation in homogeneous and heterogeneous spaces
Presenter: Hua Zuo
Abstract:
Domain adaptation aims to leverage knowledge acquired from a related domain (called source domain) to improve the efficiency of completing a prediction task (classification or regression) in the current domain (called target domain) which has different probability distribution from the source domain. We proposed fuzzy rule-based methods to deal with the domain adaptation in homogeneous and heterogeneous spaces, no matter the numbers of fuzzy rules in two domains are match or not. Further, we consider the cases that multiple source domains are available in transfer learning based on fuzzy rule-based models. Although some algorithms have been proposed, there are still issues that haven't been solved.
Location: CB11.05.100