An Adaptive Strategy for Active Learning with Smooth Decision Boundary

Andrea Locatelli, Alexandra Carpentier, Samory Kpotufe
Proceedings of Algorithmic Learning Theory, PMLR 83:547-571, 2018.

Abstract

We present the first adaptive strategy for active learning in the setting of classification with smooth decision boundary. The problem of adaptivity (to unknown distributional parameters) has remained opened since the seminal work of Castro and Nowak (2007), which first established (active learning) rates for this setting. While some recent advances on this problem establish \emph{adaptive} rates in the case of univariate data, adaptivity in the more practical setting of multivariate data has so far remained elusive. Combining insights from various recent works, we show that, for the multivariate case, a careful reduction to univariate-adaptive strategies yield near-optimal rates without prior knowledge of distributional parameters.

Cite this Paper

BibTeX
@InProceedings{pmlr-v83-locatelli18a, title = {An Adaptive Strategy for Active Learning with Smooth Decision Boundary}, author = {Locatelli, Andrea and Carpentier, Alexandra and Kpotufe, Samory}, booktitle = {Proceedings of Algorithmic Learning Theory}, pages = {547--571}, year = {2018}, editor = {Janoos, Firdaus and Mohri, Mehryar and Sridharan, Karthik}, volume = {83}, series = {Proceedings of Machine Learning Research}, month = {07--09 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v83/locatelli18a/locatelli18a.pdf}, url = {https://proceedings.mlr.press/v83/locatelli18a.html}, abstract = {We present the first adaptive strategy for active learning in the setting of classification with smooth decision boundary. The problem of adaptivity (to unknown distributional parameters) has remained opened since the seminal work of Castro and Nowak (2007), which first established (active learning) rates for this setting. While some recent advances on this problem establish \emph{adaptive} rates in the case of univariate data, adaptivity in the more practical setting of multivariate data has so far remained elusive. Combining insights from various recent works, we show that, for the multivariate case, a careful reduction to univariate-adaptive strategies yield near-optimal rates without prior knowledge of distributional parameters.} }
Endnote
%0 Conference Paper %T An Adaptive Strategy for Active Learning with Smooth Decision Boundary %A Andrea Locatelli %A Alexandra Carpentier %A Samory Kpotufe %B Proceedings of Algorithmic Learning Theory %C Proceedings of Machine Learning Research %D 2018 %E Firdaus Janoos %E Mehryar Mohri %E Karthik Sridharan %F pmlr-v83-locatelli18a %I PMLR %P 547--571 %U https://proceedings.mlr.press/v83/locatelli18a.html %V 83 %X We present the first adaptive strategy for active learning in the setting of classification with smooth decision boundary. The problem of adaptivity (to unknown distributional parameters) has remained opened since the seminal work of Castro and Nowak (2007), which first established (active learning) rates for this setting. While some recent advances on this problem establish \emph{adaptive} rates in the case of univariate data, adaptivity in the more practical setting of multivariate data has so far remained elusive. Combining insights from various recent works, we show that, for the multivariate case, a careful reduction to univariate-adaptive strategies yield near-optimal rates without prior knowledge of distributional parameters.
APA
Locatelli, A., Carpentier, A. & Kpotufe, S.. (2018). An Adaptive Strategy for Active Learning with Smooth Decision Boundary. Proceedings of Algorithmic Learning Theory, in Proceedings of Machine Learning Research 83:547-571 Available from https://proceedings.mlr.press/v83/locatelli18a.html.

Related Material