Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels
Machine Learning
2016-05-31 v6 Statistics Theory
Statistics Theory
Abstract
Crowdsourcing has become a primary means for label collection in many real-world machine learning applications. A classical method for inferring the true labels from the noisy labels provided by crowdsourcing workers is Dawid-Skene estimator. In this paper, we prove convergence rates of a projected EM algorithm for the Dawid-Skene estimator. The revealed exponent in the rate of convergence is shown to be optimal via a lower bound argument. Our work resolves the long standing issue of whether Dawid-Skene estimator has sound theoretical guarantees besides its good performance observed in practice. In addition, a comparative study with majority voting illustrates both advantages and pitfalls of the Dawid-Skene estimator.
Keywords
Cite
@article{arxiv.1310.5764,
title = {Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels},
author = {Chao Gao and Dengyong Zhou},
journal= {arXiv preprint arXiv:1310.5764},
year = {2016}
}