Corpus ID: 49554472

# Local moment matching: A unified methodology for symmetric functional estimation and distribution estimation under Wasserstein distance

@inproceedings{Han2018LocalMM,
title={Local moment matching: A unified methodology for symmetric functional estimation and distribution estimation under Wasserstein distance},
author={Y. Han and J. Jiao and T. Weissman},
booktitle={COLT},
year={2018}
}
• Published in COLT 2018
• Mathematics, Computer Science
• We present \emph{Local Moment Matching (LMM)}, a unified methodology for symmetric functional estimation and distribution estimation under Wasserstein distance. We construct an efficiently computable estimator that achieves the minimax rates in estimating the distribution up to permutation, and show that the plug-in approach of our unlabeled distribution estimator is "universal" in estimating symmetric functionals of discrete distributions. Instead of doing best polynomial approximation… CONTINUE READING
35 Citations

#### References

SHOWING 1-10 OF 56 REFERENCES
Minimax estimation of discrete distributions
• Mathematics, Computer Science
• 2015 IEEE International Symposium on Information Theory (ISIT)
• 2015
• 12
Minimax Estimation of Functionals of Discrete Distributions
• Mathematics, Computer Science
• IEEE Transactions on Information Theory
• 2015
• 193
• PDF
A Unified Maximum Likelihood Approach for Optimal Distribution Property Estimation
• Mathematics, Computer Science
• Electron. Colloquium Comput. Complex.
• 2016
• 13
• Highly Influential
• PDF
Minimax estimation of the L1 distance
• Mathematics, Computer Science
• 2016 IEEE International Symposium on Information Theory (ISIT)
• 2016
• 55
Optimal rates of entropy estimation over Lipschitz balls
• Mathematics, Computer Science
• ArXiv
• 2017
• 40
• PDF
Approximate Profile Maximum Likelihood
• Computer Science, Mathematics
• J. Mach. Learn. Res.
• 2019
• 35
• PDF
The Power of Linear Estimators
• Mathematics, Computer Science
• 2011 IEEE 52nd Annual Symposium on Foundations of Computer Science
• 2011
• 136
• Highly Influential
• PDF