Do Outliers Ruin Collaboration?

Mingda Qiao
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:4180-4187, 2018.

Abstract

We consider the problem of learning a binary classifier from $n$ different data sources, among which at most an $\eta$ fraction are adversarial. The overhead is defined as the ratio between the sample complexity of learning in this setting and that of learning the same hypothesis class on a single data distribution. We present an algorithm that achieves an $O(\eta n + \ln n)$ overhead, which is proved to be worst-case optimal. We also discuss the potential challenges to the design of a computationally efficient learning algorithm with a small overhead.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-qiao18a, title = {Do Outliers Ruin Collaboration?}, author = {Qiao, Mingda}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {4180--4187}, 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/qiao18a/qiao18a.pdf}, url = {http://proceedings.mlr.press/v80/qiao18a.html}, abstract = {We consider the problem of learning a binary classifier from $n$ different data sources, among which at most an $\eta$ fraction are adversarial. The overhead is defined as the ratio between the sample complexity of learning in this setting and that of learning the same hypothesis class on a single data distribution. We present an algorithm that achieves an $O(\eta n + \ln n)$ overhead, which is proved to be worst-case optimal. We also discuss the potential challenges to the design of a computationally efficient learning algorithm with a small overhead.} }
Endnote
%0 Conference Paper %T Do Outliers Ruin Collaboration? %A Mingda Qiao %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-qiao18a %I PMLR %P 4180--4187 %U http://proceedings.mlr.press/v80/qiao18a.html %V 80 %X We consider the problem of learning a binary classifier from $n$ different data sources, among which at most an $\eta$ fraction are adversarial. The overhead is defined as the ratio between the sample complexity of learning in this setting and that of learning the same hypothesis class on a single data distribution. We present an algorithm that achieves an $O(\eta n + \ln n)$ overhead, which is proved to be worst-case optimal. We also discuss the potential challenges to the design of a computationally efficient learning algorithm with a small overhead.
APA
Qiao, M.. (2018). Do Outliers Ruin Collaboration?. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:4180-4187 Available from http://proceedings.mlr.press/v80/qiao18a.html.

Related Material