A Robust Zero-Sum Game Framework for Pool-based Active Learning

Dixian Zhu, Zhe Li, Xiaoyu Wang, Boqing Gong, Tianbao Yang
Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics, PMLR 89:517-526, 2019.

Abstract

In this paper, we present a novel robust zero- sum game framework for pool-based active learning grounded on advanced statistical learning theory. Pool-based active learning usually consists of two components, namely, learning of a classifier given labeled data and querying of unlabeled data for labeling. Most previous studies on active learning consider these as two separate tasks and propose various heuristics for selecting important unlabeled data for labeling, which may render the selection of unlabeled examples sub-optimal for minimizing the classification error. In contrast, the present work formulates active learning as a unified optimization framework for learning the classifier, i.e., the querying of labels and the learning of models are unified to minimize a common objective for statistical learning. In addition, the proposed method avoids the issues of many previous algorithms such as inefficiency, sampling bias and sensitivity to imbalanced data distribution. Besides theoretical analysis, we conduct extensive experiments on benchmark datasets and demonstrate the superior performance of the proposed active learning method compared with the state-of-the-art methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v89-zhu19a, title = {A Robust Zero-Sum Game Framework for Pool-based Active Learning}, author = {Zhu, Dixian and Li, Zhe and Wang, Xiaoyu and Gong, Boqing and Yang, Tianbao}, booktitle = {Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics}, pages = {517--526}, year = {2019}, editor = {Chaudhuri, Kamalika and Sugiyama, Masashi}, volume = {89}, series = {Proceedings of Machine Learning Research}, month = {16--18 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v89/zhu19a/zhu19a.pdf}, url = {https://proceedings.mlr.press/v89/zhu19a.html}, abstract = {In this paper, we present a novel robust zero- sum game framework for pool-based active learning grounded on advanced statistical learning theory. Pool-based active learning usually consists of two components, namely, learning of a classifier given labeled data and querying of unlabeled data for labeling. Most previous studies on active learning consider these as two separate tasks and propose various heuristics for selecting important unlabeled data for labeling, which may render the selection of unlabeled examples sub-optimal for minimizing the classification error. In contrast, the present work formulates active learning as a unified optimization framework for learning the classifier, i.e., the querying of labels and the learning of models are unified to minimize a common objective for statistical learning. In addition, the proposed method avoids the issues of many previous algorithms such as inefficiency, sampling bias and sensitivity to imbalanced data distribution. Besides theoretical analysis, we conduct extensive experiments on benchmark datasets and demonstrate the superior performance of the proposed active learning method compared with the state-of-the-art methods.} }
Endnote
%0 Conference Paper %T A Robust Zero-Sum Game Framework for Pool-based Active Learning %A Dixian Zhu %A Zhe Li %A Xiaoyu Wang %A Boqing Gong %A Tianbao Yang %B Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Masashi Sugiyama %F pmlr-v89-zhu19a %I PMLR %P 517--526 %U https://proceedings.mlr.press/v89/zhu19a.html %V 89 %X In this paper, we present a novel robust zero- sum game framework for pool-based active learning grounded on advanced statistical learning theory. Pool-based active learning usually consists of two components, namely, learning of a classifier given labeled data and querying of unlabeled data for labeling. Most previous studies on active learning consider these as two separate tasks and propose various heuristics for selecting important unlabeled data for labeling, which may render the selection of unlabeled examples sub-optimal for minimizing the classification error. In contrast, the present work formulates active learning as a unified optimization framework for learning the classifier, i.e., the querying of labels and the learning of models are unified to minimize a common objective for statistical learning. In addition, the proposed method avoids the issues of many previous algorithms such as inefficiency, sampling bias and sensitivity to imbalanced data distribution. Besides theoretical analysis, we conduct extensive experiments on benchmark datasets and demonstrate the superior performance of the proposed active learning method compared with the state-of-the-art methods.
APA
Zhu, D., Li, Z., Wang, X., Gong, B. & Yang, T.. (2019). A Robust Zero-Sum Game Framework for Pool-based Active Learning. Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 89:517-526 Available from https://proceedings.mlr.press/v89/zhu19a.html.

Related Material