TY - JOUR
T1 - Confidence Weighted Multitask Learning
AU - Yang, Peng
AU - Zhao, Peilin
AU - Zhou, Jiayu
AU - Gao, Xin
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledged KAUST grant number(s): URF/1/3007-01-01
Acknowledgements: The research reported in this publication was supported by funding from King Abdullah University of Science and Technology (KAUST), under award number URF/1/3007-01-01.
PY - 2019/8/29
Y1 - 2019/8/29
N2 - Traditional online multitask learning only utilizes the firstorder information of the datastream. To remedy this issue, we propose a confidence weighted multitask learning algorithm, which maintains a Gaussian distribution over each task model to guide online learning process. The mean (covariance) of the Gaussian Distribution is a sum of a local component and a global component that is shared among all the tasks. In addition, this paper also addresses the challenge of active learning on the online multitask setting. Instead of requiring labels of all the instances, the proposed algorithm determines whether the learner should acquire a label by considering the confidence from its related tasks over label prediction. Theoretical results show the regret bounds can be significantly reduced. Empirical results demonstrate that the proposed algorithm is able to achieve promising learning efficacy, while simultaneously minimizing the labeling cost.
AB - Traditional online multitask learning only utilizes the firstorder information of the datastream. To remedy this issue, we propose a confidence weighted multitask learning algorithm, which maintains a Gaussian distribution over each task model to guide online learning process. The mean (covariance) of the Gaussian Distribution is a sum of a local component and a global component that is shared among all the tasks. In addition, this paper also addresses the challenge of active learning on the online multitask setting. Instead of requiring labels of all the instances, the proposed algorithm determines whether the learner should acquire a label by considering the confidence from its related tasks over label prediction. Theoretical results show the regret bounds can be significantly reduced. Empirical results demonstrate that the proposed algorithm is able to achieve promising learning efficacy, while simultaneously minimizing the labeling cost.
UR - http://hdl.handle.net/10754/630865
UR - https://aiide.org/ojs/index.php/AAAI/article/view/4507
U2 - 10.1609/aaai.v33i01.33015636
DO - 10.1609/aaai.v33i01.33015636
M3 - Article
SN - 2374-3468
VL - 33
SP - 5636
EP - 5643
JO - Proceedings of the AAAI Conference on Artificial Intelligence
JF - Proceedings of the AAAI Conference on Artificial Intelligence
ER -