TY - JOUR
T1 - Learning from Multiple Imperfect Instructors in Sensor Networks
AU - Virani, Nurali
AU - Phoha, Shashi
AU - Ray, Asok
N1 - Funding Information:
Manuscript received October 18, 2016; revised March 19, 2017 and November 3, 2017; accepted January 3, 2018. Date of publication February 1, 2018; date of current version September 17, 2018. This work was supported by the U.S. Air Force Office of Scientific Research under Grant FA9550-12-1-0270 and Grant FA9550-15-1-0400. Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the sponsoring agencies. (Corresponding author: Asok Ray.) N. Virani was with the Department of Mechanical and Nuclear Engineering, The Pennsylvania State University, University Park, PA 16801 USA. He is now with GE Global Research, Niskayuna, NY 12309 USA (e-mail: [email protected]).
Publisher Copyright:
© 2018 IEEE.
PY - 2018/10
Y1 - 2018/10
N2 - This paper presents a sequential learning framework for sensors in a network, where a few sensors assume the role of an instructor to train other sensors in the network. The instructors provide estimated labels for measurements of new sensors. These labels are possibly noisy, because a classifier of the instructor may not be perfect. A recursive density estimator is proposed to obtain the true measurement model (i.e., the observation density conditioned on the label) in spite of the training with noisy labels. Specifically, this paper answers the question "Can a sensor train other sensors?", provides necessary conditions for sensors to act as instructors, presents a sequential learning framework using recursive nonparametric kernel density estimation, and provides a convergence rate for the expected error in an observation density. The underlying concepts are illustrated and validated with simulation results.
AB - This paper presents a sequential learning framework for sensors in a network, where a few sensors assume the role of an instructor to train other sensors in the network. The instructors provide estimated labels for measurements of new sensors. These labels are possibly noisy, because a classifier of the instructor may not be perfect. A recursive density estimator is proposed to obtain the true measurement model (i.e., the observation density conditioned on the label) in spite of the training with noisy labels. Specifically, this paper answers the question "Can a sensor train other sensors?", provides necessary conditions for sensors to act as instructors, presents a sequential learning framework using recursive nonparametric kernel density estimation, and provides a convergence rate for the expected error in an observation density. The underlying concepts are illustrated and validated with simulation results.
UR - http://www.scopus.com/inward/record.url?scp=85041678360&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85041678360&partnerID=8YFLogxK
U2 - 10.1109/TNNLS.2018.2791898
DO - 10.1109/TNNLS.2018.2791898
M3 - Article
C2 - 29994429
AN - SCOPUS:85041678360
SN - 2162-237X
VL - 29
SP - 5166
EP - 5172
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
IS - 10
M1 - 8278847
ER -