Mixture Density Network Estimation of Continuous Variable Maximum Likelihood Using Discrete Training Samples
Data Analysis, Statistics and Probability
2021-08-18 v2 Machine Learning
High Energy Physics - Experiment
Abstract
Mixture Density Networks (MDNs) can be used to generate probability density functions of model parameters given a set of observables . In some applications, training data are available only for discrete values of a continuous parameter . In such situations a number of performance-limiting issues arise which can result in biased estimates. We demonstrate the usage of MDNs for parameter estimation, discuss the origins of the biases, and propose a corrective method for each issue.
Cite
@article{arxiv.2103.13416,
title = {Mixture Density Network Estimation of Continuous Variable Maximum Likelihood Using Discrete Training Samples},
author = {Charles Burton and Spencer Stubbs and Peter Onyisi},
journal= {arXiv preprint arXiv:2103.13416},
year = {2021}
}