Title :
A neural-network Q-learning method for decentralized sequential detection with feedback
Author :
Guo, Chengan ; Kuh, Anthony
Author_Institution :
Dept. of Electr. Eng., Hawaii Univ., Honolulu, HI, USA
fDate :
6/21/1905 12:00:00 AM
Abstract :
This paper studies a feedback decentralized sequential detection system using Q-learning. The purpose is to obtain a better understanding of certain distributed detection systems and to examine the impact of feedback information. Performance comparisons are made in the paper between the Q-learning approach, the centralized sequential probability ratio test method, a dynamic programming method, and a non-feedback distributed detection method
Keywords :
feedback; learning (artificial intelligence); neural nets; sensor fusion; signal detection; Q-learning; decentralized sequential detection; distributed detection systems; feedback; neural-network; reinforcement learning; sensor fusion; Bayesian methods; Cost function; Detectors; Distributed processing; Dynamic programming; Neural networks; Neurofeedback; Reliability theory; Sensor systems; Sequential analysis;
Conference_Titel :
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Print_ISBN :
0-7803-5529-6
DOI :
10.1109/IJCNN.1999.833419
