Iterative Teaching by Data Hallucination (original) (raw)
[edit]
Zeju Qiu, Weiyang Liu, Tim Z. Xiao, Zhen Liu, Umang Bhatt, Yucen Luo, Adrian Weller, Bernhard Schölkopf
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:9892-9913, 2023.
Abstract
We consider the problem of iterative machine teaching, where a teacher sequentially provides examples based on the status of a learner under a discrete input space (i.e., a pool of finite samples), which greatly limits the teacher’s capability. To address this issue, we study iterative teaching under a continuous input space where the input example (i.e., image) can be either generated by solving an optimization problem or drawn directly from a continuous distribution. Specifically, we propose data hallucination teaching (DHT) where the teacher can generate input data intelligently based on labels, the learner’s status and the target concept. We study a number of challenging teaching setups (e.g., linear/neural learners in omniscient and black-box settings). Extensive empirical results verify the effectiveness of DHT.
Cite this Paper
BibTeX
@InProceedings{pmlr-v206-qiu23a, title = {Iterative Teaching by Data Hallucination}, author = {Qiu, Zeju and Liu, Weiyang and Xiao, Tim Z. and Liu, Zhen and Bhatt, Umang and Luo, Yucen and Weller, Adrian and Sch\"olkopf, Bernhard}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {9892--9913}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/qiu23a/qiu23a.pdf}, url = {https://proceedings.mlr.press/v206/qiu23a.html}, abstract = {We consider the problem of iterative machine teaching, where a teacher sequentially provides examples based on the status of a learner under a discrete input space (i.e., a pool of finite samples), which greatly limits the teacher’s capability. To address this issue, we study iterative teaching under a continuous input space where the input example (i.e., image) can be either generated by solving an optimization problem or drawn directly from a continuous distribution. Specifically, we propose data hallucination teaching (DHT) where the teacher can generate input data intelligently based on labels, the learner’s status and the target concept. We study a number of challenging teaching setups (e.g., linear/neural learners in omniscient and black-box settings). Extensive empirical results verify the effectiveness of DHT.} }
Endnote
%0 Conference Paper %T Iterative Teaching by Data Hallucination %A Zeju Qiu %A Weiyang Liu %A Tim Z. Xiao %A Zhen Liu %A Umang Bhatt %A Yucen Luo %A Adrian Weller %A Bernhard Schölkopf %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-qiu23a %I PMLR %P 9892--9913 %U https://proceedings.mlr.press/v206/qiu23a.html %V 206 %X We consider the problem of iterative machine teaching, where a teacher sequentially provides examples based on the status of a learner under a discrete input space (i.e., a pool of finite samples), which greatly limits the teacher’s capability. To address this issue, we study iterative teaching under a continuous input space where the input example (i.e., image) can be either generated by solving an optimization problem or drawn directly from a continuous distribution. Specifically, we propose data hallucination teaching (DHT) where the teacher can generate input data intelligently based on labels, the learner’s status and the target concept. We study a number of challenging teaching setups (e.g., linear/neural learners in omniscient and black-box settings). Extensive empirical results verify the effectiveness of DHT.
APA
Qiu, Z., Liu, W., Xiao, T.Z., Liu, Z., Bhatt, U., Luo, Y., Weller, A. & Schölkopf, B.. (2023). Iterative Teaching by Data Hallucination. _Proceedings of The 26th International Conference on Artificial Intelligence and Statistics_, in _Proceedings of Machine Learning Research_ 206:9892-9913 Available from https://proceedings.mlr.press/v206/qiu23a.html.