ICML'96 Paper Available

[Mance E. Harmon]

The following paper will be presented at the 13th International Conference on 
Machine Learning, Bari, Italy, 3-6 July, and is now available in postscript 
and RTF formats at the following URL:

http://www.aa.wpafb.af.mil/~harmonme


Residual Q-Learning Applied to Visual Attention


Cesar Bandera
Amherst Systems, Inc.
Machine Vision Dept.30 Wilson Road
Buffalo, New York14221-7082
cba at amherst.com	

Francisco J. Vico,Jose M. Bravo
Facultad de Psicologia
Universidad de Malaga
29017 Malaga (Spain)
fjv at eva.psi.uma.es
jbm at eva.psi.uma.es	

Mance E. Harmon
Wright Laboratory
WL/AACF 
2241 Avionics Circle
Wright-Patterson AFB,Ohio  45433-7318
harmonme at aa.wpafb.af.mil	

Leemon C. Baird III
U.S.A.F. Academy
2354 Fairchild Dr.
Suite 6K41
USAFA, Colorado 80840-6234
baird at cs.usafa.af.mil


ABSTRACT

Foveal vision features imagers with graded acuity coupled with context 
sensitive sensor gaze control, analogous to that prevalent throughout 
vertebrate vision.  Foveal vision operates more efficiently than uniform 
acuity vision because resolution is treated as a dynamically allocatable 
resource, but requires a more refined visual attention mechanism.  We 
demonstrate that reinforcement learning (RL) significantly improves the 
performance of foveal visual attention, and of the overall vision system, for 
the task of model based target recognition.  A simulated foveal vision system 
is shown to classify targets with fewer fixations by learning strategies for 
the acquisition of visual information relevant to the task, and learning how 
to generalize these strategies in ambiguous and unexpected scenario 
conditions.