Negative Results for Active Learning with Convex Losses

Steve Hanneke, Liu Yang
; Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, JMLR Workshop and Conference Proceedings 9:321-325, 2010.

Abstract

We study the problem of active learning with convex loss functions. We prove that even under bounded noise constraints, the minimax rates for proper active learning are often no better than passive learning.

Cite this Paper


BibTeX
@InProceedings{pmlr-v9-hanneke10a, title = {Negative Results for Active Learning with Convex Losses}, author = {Steve Hanneke and Liu Yang}, booktitle = {Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics}, pages = {321--325}, year = {2010}, editor = {Yee Whye Teh and Mike Titterington}, volume = {9}, series = {Proceedings of Machine Learning Research}, address = {Chia Laguna Resort, Sardinia, Italy}, month = {13--15 May}, publisher = {JMLR Workshop and Conference Proceedings}, pdf = {http://proceedings.mlr.press/v9/hanneke10a/hanneke10a.pdf}, url = {http://proceedings.mlr.press/v9/hanneke10a.html}, abstract = {We study the problem of active learning with convex loss functions. We prove that even under bounded noise constraints, the minimax rates for proper active learning are often no better than passive learning.} }
Endnote
%0 Conference Paper %T Negative Results for Active Learning with Convex Losses %A Steve Hanneke %A Liu Yang %B Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2010 %E Yee Whye Teh %E Mike Titterington %F pmlr-v9-hanneke10a %I PMLR %J Proceedings of Machine Learning Research %P 321--325 %U http://proceedings.mlr.press %V 9 %W PMLR %X We study the problem of active learning with convex loss functions. We prove that even under bounded noise constraints, the minimax rates for proper active learning are often no better than passive learning.
RIS
TY - CPAPER TI - Negative Results for Active Learning with Convex Losses AU - Steve Hanneke AU - Liu Yang BT - Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics PY - 2010/03/31 DA - 2010/03/31 ED - Yee Whye Teh ED - Mike Titterington ID - pmlr-v9-hanneke10a PB - PMLR SP - 321 DP - PMLR EP - 325 L1 - http://proceedings.mlr.press/v9/hanneke10a/hanneke10a.pdf UR - http://proceedings.mlr.press/v9/hanneke10a.html AB - We study the problem of active learning with convex loss functions. We prove that even under bounded noise constraints, the minimax rates for proper active learning are often no better than passive learning. ER -
APA
Hanneke, S. & Yang, L.. (2010). Negative Results for Active Learning with Convex Losses. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, in PMLR 9:321-325

Related Material