Domain Adaptation Meets Active Learning
Request pdf domain adaptation meets active learning in this work we show how active learning in some target domain can leverage information from a different but related source domain.
Domain adaptation meets active learning. In this work we show how active learning in some target domain can leverage information from a different but related source domain. By hao lu 1 lei zhang 2 zhiguo cao 1 wei wei 2 ke xian 1 chunhua shen 3 anton van den hengel 3. Active learning aims to minimize labeling effort by selecting the most informative.
Tensor aligned invariant subspace learning. Ieee international conference on computer vision iccv 2017. Alnlp ws sig.
1 huazhong university of science and technology china. Los angeles california venues. By piyush rai avishek saha hal daumé iii and suresh venkatasubramanian.
In this paper we apply active learning strategies to domain adaptation for named entity recognition systems and show that adaptive learning combining the source and target domains is more effective than non adaptive learning directly from the target domain. Piyush rai avishek saha hal daumé suresh venkatasubramanian. Domain adaptation meets active learning.
Domain adaptation meets active learning piyush rai avishek saha hal daum e iii and suresh venkatasubramanian school of computing university of utah salt lake city ut 84112 fpiyush avishek hal suresh g cs utah edu abstract in this work we show how active learning in some target domain can leverage infor. When unsupervised domain adaptation meets tensor representations. Domain adaptation meets active learning piyush rai avishek saha hal daum e iii and suresh venkatasubramanian school of computing university of utah salt lake city ut 84112 piyush avishek hal suresh cs utah edu abstract in this work we show how active learning in some target domain can leverage infor mation from a different but.
Proceedings of the naacl hlt 2010 workshop on active learning for natural language processing month. In this work we show how active learning in some target domain can leverage information from a different but related source domain. We present an algorithm that harnesses the source domain data to learn the best possible initializer.