TY - GEN
T1 - Improved finite-sample estimate of a nonparametric f-divergence
AU - Kadambi, Prad
AU - Wisler, Alan
AU - Berisha, Visar
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/2
Y1 - 2017/7/2
N2 - Information divergence functions allow us to measure distances between probability density functions. We focus on the case where we only have data from the two distributions and have no knowledge of the underlying models from which the data is sampled. In this scenario, we consider an f-divergence for which there exists an asymptotically consistent, nonparametric estimator based on minimum spanning trees, the Dp divergence. Nonparametric estimators are known to have slow convergence rates in higher dimensions (d > 4), resulting in a large bias for small datasets. Based on experimental validation, we conjecture that the original estimator follows a power law convergence model and introduce a new estimator based on a bootstrap sampling scheme that results in a reduced bias. Experiments on real and artificial data show that the new estimator results in improved estimates of the Dp divergence when compared against the original estimator.
AB - Information divergence functions allow us to measure distances between probability density functions. We focus on the case where we only have data from the two distributions and have no knowledge of the underlying models from which the data is sampled. In this scenario, we consider an f-divergence for which there exists an asymptotically consistent, nonparametric estimator based on minimum spanning trees, the Dp divergence. Nonparametric estimators are known to have slow convergence rates in higher dimensions (d > 4), resulting in a large bias for small datasets. Based on experimental validation, we conjecture that the original estimator follows a power law convergence model and introduce a new estimator based on a bootstrap sampling scheme that results in a reduced bias. Experiments on real and artificial data show that the new estimator results in improved estimates of the Dp divergence when compared against the original estimator.
KW - asymptotic estimator
KW - bootstrap estimator
KW - minimal graphs
KW - nonparametric f-Divergence
UR - http://www.scopus.com/inward/record.url?scp=85050914751&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85050914751&partnerID=8YFLogxK
U2 - 10.1109/ACSSC.2017.8335474
DO - 10.1109/ACSSC.2017.8335474
M3 - Conference contribution
AN - SCOPUS:85050914751
T3 - Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017
SP - 875
EP - 879
BT - Conference Record of 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017
A2 - Matthews, Michael B.
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 51st Asilomar Conference on Signals, Systems and Computers, ACSSC 2017
Y2 - 29 October 2017 through 1 November 2017
ER -