Noise-contrastive estimation: A new estimation principle for unnormalized statistical models Michael Gutmann Dept of Computer Science and HIIT, University of Helsinki

First Page		Document Content
Date: 2018-07-14 18:55:46 Estimation theory Econometrics Statistical inference Estimator Probability distribution fitting M-estimators Maximum likelihood estimation Fisher information Gamma distribution Maximum spacing estimation		Noise-contrastive estimation: A new estimation principle for unnormalized statistical models Michael Gutmann Dept of Computer Science and HIIT, University of Helsinki Add to Reading List Source URL: proceedings.mlr.press Download Document from Source Website File Size: 618,78 KB Share Document on Facebook

	1 Citation count distributions for large monodisciplinary journals1 Mike Thelwall, Statistical Cybermetrics Research Group, University of Wolverhampton, UK. Many different citation-based indicators are used by researche DocID: 1qPVN - View Document
	COMBINED ALGORITHMS FOR FITTING FINITE MIXTURE DISTRIBUTIONS COMBINED ALGORITHMS FOR CONSTRAINED ESTIMATION OF FINITE MIXTURE DISTRIBUTIONS WITH GROUPED DocID: 1p3jl - View Document
	Fitting Power Law Distributions to Data Willy Lai Introduction In this paper, we will be testing whether the frequency of family names from the 2000 Census follow a power law distribution. Power law distributions are usu DocID: 1oRf1 - View Document
	1 Are the discretised lognormal and hooked power law distributions plausible for citation data? Mike Thelwall, Statistical Cybermetrics Research Group, University of Wolverhampton, UK. There is no agreement over which s DocID: 1otNc - View Document
	Fitting Distributions to Data Practical Issues in the Use of Probabilistic Risk Assessment Sarasota, Florida February 28 - March 2, 1999 DocID: 1gAl3 - View Document