Towards Black-box Iterative Machine Teaching

Weiyang Liu, Bo Dai, Xingguo Li, Zhen Liu, James Rehg, Le Song
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:3141-3149, 2018.

Abstract

In this paper, we make an important step towards the black-box machine teaching by considering the cross-space machine teaching, where the teacher and the learner use different feature representations and the teacher can not fully observe the learner’s model. In such scenario, we study how the teacher is still able to teach the learner to achieve faster convergence rate than the traditional passive learning. We propose an active teacher model that can actively query the learner (i.e., make the learner take exams) for estimating the learner’s status and provably guide the learner to achieve faster convergence. The sample complexities for both teaching and query are provided. In the experiments, we compare the proposed active teacher with the omniscient teacher and verify the effectiveness of the active teacher model.

Cite this Paper

BibTeX
@InProceedings{pmlr-v80-liu18b, title = {Towards Black-box Iterative Machine Teaching}, author = {Liu, Weiyang and Dai, Bo and Li, Xingguo and Liu, Zhen and Rehg, James and Song, Le}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {3141--3149}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/liu18b/liu18b.pdf}, url = {https://proceedings.mlr.press/v80/liu18b.html}, abstract = {In this paper, we make an important step towards the black-box machine teaching by considering the cross-space machine teaching, where the teacher and the learner use different feature representations and the teacher can not fully observe the learner’s model. In such scenario, we study how the teacher is still able to teach the learner to achieve faster convergence rate than the traditional passive learning. We propose an active teacher model that can actively query the learner (i.e., make the learner take exams) for estimating the learner’s status and provably guide the learner to achieve faster convergence. The sample complexities for both teaching and query are provided. In the experiments, we compare the proposed active teacher with the omniscient teacher and verify the effectiveness of the active teacher model.} }
Endnote
%0 Conference Paper %T Towards Black-box Iterative Machine Teaching %A Weiyang Liu %A Bo Dai %A Xingguo Li %A Zhen Liu %A James Rehg %A Le Song %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-liu18b %I PMLR %P 3141--3149 %U https://proceedings.mlr.press/v80/liu18b.html %V 80 %X In this paper, we make an important step towards the black-box machine teaching by considering the cross-space machine teaching, where the teacher and the learner use different feature representations and the teacher can not fully observe the learner’s model. In such scenario, we study how the teacher is still able to teach the learner to achieve faster convergence rate than the traditional passive learning. We propose an active teacher model that can actively query the learner (i.e., make the learner take exams) for estimating the learner’s status and provably guide the learner to achieve faster convergence. The sample complexities for both teaching and query are provided. In the experiments, we compare the proposed active teacher with the omniscient teacher and verify the effectiveness of the active teacher model.
APA
Liu, W., Dai, B., Li, X., Liu, Z., Rehg, J. & Song, L.. (2018). Towards Black-box Iterative Machine Teaching. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:3141-3149 Available from https://proceedings.mlr.press/v80/liu18b.html.

Related Material