|
Activity Number:
|
610
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 3, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Defense and National Security
|
|
Abstract #318967
|
View Presentation
|
|
Title:
|
Modeling the Upper Tail of the Distribution of Facial Recognition Non-Match Scores
|
|
Author(s):
|
Brett Hunter* and Dan Cooley and Geof Givens and J. Ross Beveridge
|
|
Companies:
|
and Colorado State University and Colorado State University and Colorado State University
|
|
Keywords:
|
M-estimation ;
generalized Pareto ;
quantile regression
|
|
Abstract:
|
In facial recognition applications, the upper tail of the distribution of non-match scores is of interest because existing algorithms classify a pair of images as a match if their score exceeds some high quantile of the non-match distribution. We construct a general model for the distribution above the (1-?)th quantile borrowing ideas from extreme value theory. Our resulting distribution can be viewed as a reparameterized generalized Pareto distribution (GPD), but it differs from the traditional GPD in that ? is treated as fixed. Inference for both threshold and the GPD scale and shape parameters is performed via M-estimation, where our objective function is a combination of the quantile regression loss function and reparameterized GPD densities. Parameters are treated as a function of covariates in order to gain understanding of their influence on the tail of the distribution of non-match scores. Our simulation study shows that our method is able to estimate both the set of parameters describing the covariates' influence and high quantiles of the non-match distribution. We apply our method to a data set of non-match scores.
|
Authors who are presenting talks have a * after their name.
