TY - GEN
T1 - Transfer learning from multiple source domains via consensus regularization
AU - Luo, Ping
AU - Zhuang, Fuzhen
AU - Xiong, Hui
AU - Xiong, Yuhong
AU - He, Qing
PY - 2008
Y1 - 2008
N2 - Recent years have witnessed an increased interest in transfer learning. Despite the vast amount of research performed in this field, there are remaining challenges in applying the knowledge learnt from multiple source domains to a target domain. First, data from multiple source domains can be semantically related, but have different distributions. It is not clear how to exploit the distribution differences among multiple source domains to boost the learning performance in a target domain. Second, many real-world applications demand this transfer learning to be performed in a distributed manner. To meet these challenges, we propose a consensus regularization framework for transfer learning from multiple source domains to a target domain. In this framework, a local classifier is trained by considering both local data available in a source domain and the prediction consensus with the classifiers from other source domains. In addition, the training algorithm can be implemented in a distributed manner, in which all the source-domains are treated as slave nodes and the target domain is used as the master node. To combine the training results from multiple source domains, it only needs share some statistical data rather than the full contents of their labeled data. This can modestly relieve the privacy concerns and avoid the need to upload all data to a central location. Finally, our experimental results show the effectiveness of our consensus regularization learning.
AB - Recent years have witnessed an increased interest in transfer learning. Despite the vast amount of research performed in this field, there are remaining challenges in applying the knowledge learnt from multiple source domains to a target domain. First, data from multiple source domains can be semantically related, but have different distributions. It is not clear how to exploit the distribution differences among multiple source domains to boost the learning performance in a target domain. Second, many real-world applications demand this transfer learning to be performed in a distributed manner. To meet these challenges, we propose a consensus regularization framework for transfer learning from multiple source domains to a target domain. In this framework, a local classifier is trained by considering both local data available in a source domain and the prediction consensus with the classifiers from other source domains. In addition, the training algorithm can be implemented in a distributed manner, in which all the source-domains are treated as slave nodes and the target domain is used as the master node. To combine the training results from multiple source domains, it only needs share some statistical data rather than the full contents of their labeled data. This can modestly relieve the privacy concerns and avoid the need to upload all data to a central location. Finally, our experimental results show the effectiveness of our consensus regularization learning.
KW - Classification
KW - Consensus regularization
KW - Transfer learning
UR - http://www.scopus.com/inward/record.url?scp=70049089673&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70049089673&partnerID=8YFLogxK
U2 - 10.1145/1458082.1458099
DO - 10.1145/1458082.1458099
M3 - Conference contribution
AN - SCOPUS:70049089673
SN - 9781595939913
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 103
EP - 112
BT - Proceedings of the 17th ACM Conference on Information and Knowledge Management, CIKM'08
T2 - 17th ACM Conference on Information and Knowledge Management, CIKM'08
Y2 - 26 October 2008 through 30 October 2008
ER -