DocumentCode :
3535225
Title :
Online evaluation of communication models derived via transfer learning
Author :
Wagle, Neeti ; Frew, Eric W.
Author_Institution :
Dept. of Comput. Sci., Univ. of Colorado, Boulder, CO, USA
fYear :
2012
fDate :
3-7 Dec. 2012
Firstpage :
1609
Lastpage :
1613
Abstract :
This paper assesses an online method for testing the plausibility of a transferred Gaussian process (GP) communication model in a new environment based on a minimal number of signal measurements. The transferred GP is constructed by combining hyperparameters learned in the original environment with signal measurements collected by an unmanned aircraft (UA) in the new goal environment. The GP´s performance and plausibility is evaluated on a separate validation set in a quick and computationally efficient manner using decision-theoretic hypothesis testing and risk minimization.
Keywords :
Gaussian processes; aircraft communication; autonomous aerial vehicles; decision theory; risk management; GP communication model; GP performance; UA; decision-theoretic hypothesis testing; hyperparameter; online evaluation; online method; risk minimization; signal measurement; transfer learning; transferred Gaussian process; unmanned aircraft; Computational modeling; Correlation; Gaussian processes; Noise; Radio frequency; Testing; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Globecom Workshops (GC Wkshps), 2012 IEEE
Conference_Location :
Anaheim, CA
Print_ISBN :
978-1-4673-4942-0
Electronic_ISBN :
978-1-4673-4940-6
Type :
conf
DOI :
10.1109/GLOCOMW.2012.6477826
Filename :
6477826
Link To Document :
بازگشت