ICML 2020poster24 citations
Optimal Robust Learning of Discrete Distributions from Batches
Abstract
Many applications, including natural language processing, sensor networks, collaborative filtering, and federated learning, call for estimating discrete distributions from data collected in batches, some of which may be untrustworthy, erroneous, faulty, or even adversarial. Previous estimators for this setting ran in exponential time, and for some regimes required a suboptimal number of batches. We provide the first polynomial-time estimator that is optimal in the number of batches and achieves essentially the best possible estimation accuracy.
BibTeX
@InProceedings{pmlr-v119-jain20a,
title = {Optimal Robust Learning of Discrete Distributions from Batches},
author = {Jain, Ayush and Orlitsky, Alon},
booktitle = {Proceedings of the 37th International Conference on Machine Learning},
pages = {4651--4660},
year = {2020},
editor = {III, Hal Daumé and Singh, Aarti},
volume = {119},
series = {Proceedings of Machine Learning Research},
month = {13--18 Jul},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v119/jain20a/jain20a.pdf},
url = {https://proceedings.mlr.press/v119/jain20a.html},
abstract = {Many applications, including natural language processing, sensor networks, collaborative filtering, and federated learning, call for estimating discrete distributions from data collected in batches, some of which may be untrustworthy, erroneous, faulty, or even adversarial. Previous estimators for this setting ran in exponential time, and for some regimes required a suboptimal number of batches. We provide the first polynomial-time estimator that is optimal in the number of batches and achieves essentially the best possible estimation accuracy.}
}