Title :
Reinforcement learning in multiresolution object recognition
Author :
Iftekharuddin, Khan M. ; Widjanarko, Taufiq
Author_Institution :
Dept. of Electr. & Comput. Eng., Memphis Univ., USA
Abstract :
In this work, we propose an adaptive automatic target recognition (ATR) technique that exploits reinforcement learning (RL) for multiresolution object recognition. The RL structure is the implementation of neuro-dynamic programming (NDP) for the critic and action networks. The critic network calculates the cost-to-go function J* based on a simplistic ATR plant that involves multiresolution images as the input state variable. The calculation of this function, J* includes the role of the reinforcement signal in the critic network. Output of this critic stage is fed back to update the weights of both action and critic networks respectively. Our simulation results suggest that RL may be successfully integrated into an adaptive multiresolution ATR framework.
Keywords :
dynamic programming; image resolution; learning (artificial intelligence); neural nets; object recognition; action networks; adaptive automatic target recognition; adaptive multiresolution ATR; cost-to-go function; multiresolution image recognition; multiresolution object recognition; neurodynamic programming; reinforcement learning; Artificial neural networks; Biological systems; Computer hacking; Image processing; Image resolution; Intelligent systems; Learning; Object recognition; Signal resolution; Target recognition;
Conference_Titel :
Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on
Print_ISBN :
0-7803-8359-1
DOI :
10.1109/IJCNN.2004.1380085