A Caltech Library Service

Near-Optimal Machine Teaching via Explanatory Teaching Sets

Chen, Yuxin and Mac Aodha, Oisin and Su, Shihan and Perona, Pietro and Yue, Yisong (2018) Near-Optimal Machine Teaching via Explanatory Teaching Sets. Proceedings of Machine Learning Research, 84 . pp. 1970-1978. ISSN 1938-7228.

[img] PDF - Published Version
See Usage Policy.


Use this Persistent URL to link to this item:


Modern applications of machine teaching for humans often involve domain-specific, non- trivial target hypothesis classes. To facilitate understanding of the target hypothesis, it is crucial for the teaching algorithm to use examples which are interpretable to the human learner. In this paper, we propose NOTES, a principled framework for constructing interpretable teaching sets, utilizing explanations to accelerate the teaching process. Our algorithm is built upon a natural stochastic model of learners and a novel submodular surrogate objective function which greedily selects interpretable teaching examples. We prove that NOTES is competitive with the optimal explanation-based teaching strategy. We further instantiate NOTES with a specific hypothesis class, which can be viewed as an interpretable approximation of any hypothesis class, allowing us to handle complex hypothesis in practice. We demonstrate the effectiveness of NOTES on several image classification tasks, for both simulated and real human learners. Our experimental results suggest that by leveraging explanations, one can significantly speed up teaching.

Item Type:Article
Related URLs:
URLURL TypeDescription Material
Mac Aodha, Oisin0000-0002-5787-5073
Perona, Pietro0000-0002-7583-5809
Yue, Yisong0000-0001-9127-1989
Additional Information:© 2018 The Author(s). This work was supported in part by Northrop Grumman, Bloomberg, AWS Research Credits, Google as part of the Visipedia project, and a Swiss NSF Early Mobility Postdoctoral Fellowship.
Funding AgencyGrant Number
Northrop Grumman CorporationUNSPECIFIED
Amazon Web ServicesUNSPECIFIED
Swiss National Science Foundation (SNSF)UNSPECIFIED
Series Name:Proceedings of Machine Learning Research
Record Number:CaltechAUTHORS:20180613-142015680
Persistent URL:
Official Citation:Yuxin Chen, Oisin Mac Aodha, Shihan Su, Pietro Perona, Yisong Yue ; Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, PMLR 84:1970-1978, 2018.
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:87076
Deposited By: Caroline Murphy
Deposited On:13 Jun 2018 21:39
Last Modified:09 Mar 2020 13:19

Repository Staff Only: item control page